Академический Документы
Профессиональный Документы
Культура Документы
BUREAUCRACY
DILEMMAS OF
THE INDIVIDUAL
IN PUBLIC SERVICES
MICHAEL LIPSKY
Street-Level Bureaucracy
STRE ET-LE VEL
BUREAUCRACY
MICHA EL LIPSK Y
The Russell Sage Foundation, one of the oldest of America's general pmpose founda-
tions, was established in 1907 by Mrs. Margaret Olivia Sage for "the improvement of
social and living conditions in the United States." The Foundation seeks to fulfill this
mandate by fostering the development and dissemination of knowledge about the coun-
try's political, social, and economic problems. While the Foundation endeavors to assure
the accuracy and objectivity of each book it publishes, the conclusions and interpreta-
tions in Russell Sage Foundation publications are those of the authors and not of the
Foundation, its Trustees, or its staff. Publication by Russell Sage, therefore, does not
imply Foundation endorsement.
BOARD OF TRUSTEES
Mary C. Waters, Chair
Kenneth D. Brody Kathleen Hall Jamieson Shelley E. Taylor
W: Bowman Cutter, III Lawrence F. Katz Richard H. Thaler
Robert E. Denham, Esq. Melvin J. Konner Eric Wanner
John A. Ferejohn Sara S. McLanahan
Larry V. Hedges Nancy Rosenblum
Copyright 2010 by Russell Sage Foundation. All rights reserved. Printed in the United
States of America. Original edition published in 1980. No part of this publication may
be reproduced, stored in a retrieval system, or transmitted in any form or by any means,
electronic, mechanical, photocopying, recording, or otherwise, without the prior written
permission of the publisher.
Reproduction by the United States Government in whole or in part is permitted for any
purpose.
The paper used in this publication meets the minimum requirements of American Na-
tional Standard for Information Sciences-Permanence of Paper for Printed Library
Materials. ANSI Z39.48-1992.
RUSSELL SAGE FOUNDATION
112 East 64th Street, New York, New York 10065
10987654321
For Joshua and Jacob
CONTEN TS
PART I
INTRODUCTI ON
PART II
CONDITIONS OF WORK
INTRODUCTION 27
3. The Problem of Resources 29
DEMAND AND SUPPLY, OR WHY RESOURCES ARE
USUALLY INADEQUATE IN STREET-LEVEL
BUREAUCRACIES 33
4. Goals and Performance Measures 40
GOALS 40
PERFORMANCE MEASURES 48
5. Relations with Clients 54
NONVOLUNTARY CLIENTS 54
CONFLICT, RECIPROCITY, AND CONTROL 57
THE SOCIAL CONSTRUCTION OF A CLIENT 59
vii
Contents
6. Advocacy and Alienation in Street-Level Work 71
ADVOCACY 72
ALIENATION 75
IMPLICATIONS OF ALIENATION 79
PART III
PATTERNS OF PRACTICE
INTRODUCTION 81
7. Rationing Services: Limitation ofAccess and Demand 87
THE COSTS OF SERVICE 88
QUEUING 95
ROUTINES AND RATIONING 99
8. Rationing Services: Inequality in Administration 105
A COMMENT ON THE UBIQUITY OF BIAS 111
9. Controlling Clients and the Work Situation 117
HUSBANDING RESOURCES 125
MANAGING THE CONSEQUENCES OF ROUTINE
PRACTICE 133
10. The Client-Processing Mentality 140
MODIFICATIONS OF CONCEPTIONS OF WORK 142
MODIFICATIONS OF CONCEPTIONS OF CLIENTS 151
PART IV
THE FUTURE OF STREET-LEVEL
BUREAUCRACY
11. The Assault on Human Services: Bureaucratic
Control, Accountability, and the Fiscal Crisis 159
HOLDING WORKERS TO AGENCY OBJECTIVES 162
ACCOUNTABILITY AND PRODUCTIVITY 170
STREET-LEVEL BUREAUCRATS AND THE FISCAL
CRISIS 172
12. The Broader Context of Bureaucratic Relations 180
CONTRADICTORY TENDENCIES IN STREET-LEVEL
BUREAUCRATIC RELATIONS 188
viii
Contents
13. Support for Human Services: Notes for Reform and
Reconstruction
DIRECTIONS FOR GREATER CLIENT AUTONOMY
DIRECTIONS FOR CURRENT PRACTICE
THE PROSPECTS AND PROBLEMS OF
PROFESSIONAL ISM 201
KEEPING NEW PROFESSIONAL S NEW 204
NOTES 239
m~ ~
ix
PREFA CE
This book is in part a search for the place of the individual in those public
services I call street-level bureaucracie s. These are the schools, police and
welfare departments , lower courts, legal services offices, and other agencies
whose workers interact with and have wide discretion over the dispensation
of benefits or the allocation of public sanctions.
It is also an inquiry into the structure of one of those resonant moments
in civic life. Like driving on interstate highways, playing in a public park,
voting, dining in a smoke-free restaurant, paying taxes, and listening to Na-
tional Public Radio, interactions with street-level bureaucracie s are places
where citizens experience directly the government they have implicitly con-
structed. Unlike these other experiences, however, citizen encounters with
street-level bureaucracie s are not straightforward; instead, they involve
complex interactions with public workers that may deeply affect the bene-
fits and sanctions they receive.
Street-Level Bureaucracy was originally published in 1g8o and made two
distinctive claims. The first was that the exercise of discretion was a critical
dimension of much of the work of teachers, social workers, police officers,
and other public workers who regularly interact with citizens in the course
of their jobs. Further, the jobs typically could not be performed according
to the highest standards of decision making in the various fields because
street-level workers lacked the time, information, or other resources neces-
sary to respond properly to the individual case. Instead, street-level bureau-
xi
Preface
crats manage their difficult jobs by developing routines of practice and psy-
chologically simplifying their clientele and environment in ways that
strongly influence the outcomes of their efforts. Mass processing of clients
is the norm, and has important implications for the quality of treatment and
services.
These observations are instructive in themselves, and have profound im-
plications for public policy. They suggest that understanding public policies
in street-level bureaucracies requires analysis of how the unsanctioned work
responses of street-level bureaucrats combine with rules and agency pro-
nouncements to add up to what the public ultimately experiences as agency
performance.
The second claim was that work as diverse and apparently unrelated as
that of guidance counselors, judges, police officers, and social workers to a
degree is structurally similar, so that one could compare these work settings
with each other. Describing front-line public service delivery in terms of a
small number of analytic characteristics made possible a new way of seeing
these very familiar public roles, and how they were like and different from
one another.
However diverse these occupations otherwise are, they could now be
seen as embodying an essential paradox that plays out in a variety of ways.
On the one hand, the work is often highly scripted to achieve policy objec-
tives that have their origins in the political process. On the other hand, the
work requires improvisation and responsiveness to the individual case. Not
only that, but generally the public wants administrators of public services to
be at least open to the possibility that a special case is presenting itself, or
that extraordinary efforts of one sort or another are called for.
Essentially all the great reform efforts of the last thirty years to improve
performance or accountability in street-level public services may be under-
stood as attempts to manage this apparently paradoxical reality: how to treat
all citizens alike in their claims on government, and how at the same time to
be responsive to the individual case when appropriate. The phrase "street-
level bureaucracy" hints at this paradox. "Bureaucracy" implies a set of rules
and structures of authority; "street-level" implies a distance from the center
where authority presumably resides.
In Street-Level Bureaucracy, I show how people experience public poli-
cies in realms that are critical to our welfare and sense of community. Too
often we read about education, policing, social work, and other vital public
services without realizing or being given concrete understanding of how
these public policies result from the aggregation of the separate actions of
xii
Dilemmas of the Individual in Public Services
many individuals, or how and why the actions in question are consistently
reproduced by the behavior of individuals.
The book is grounded in observations of the collective behavior of public
service organizations and advances a theory of the work of street-level bu-
reaucracies as individual workers experience it. I argue that the decisions of
street-level bureaucrats, the routines they establish, and the devices they
invent to cope with uncertainties and work pressures, effectively become the
public policies they carry out. I maintain that public policy is not best un-
derstood as made in legislatures or top-floor suites of high-ranking adminis-
trators. These decision-making arenas are important, of course, but they do
not represent the complete picture. To the mix of places where policies are
made, one must add the crowded offices and daily encounters of street-level
workers. Further, I point out that policy conflict is not only expressed as the
contention of interest groups, as we have come to expect. It is also located
in the struggles between individual workers and citizens who challenge or
submit to client-processing.
For example, many people are convinced that police officers dispropor-
tionately single out African Americans for scrutiny and wrongly use skin
color and racial characteristics to target blacks for attention. Police officials
invariably deny that they engage in racial profiling, and suggest that if blacks
are stopped disproportionately it is because they act in ways that legiti-
mately trigger police inquiry. It is evident that, to the extent racial profiling
exists, it arises not from official policy or direct racial orientations but out of
the ways police officers draw on social stereotypes in exercising the discre-
tion sanctioned by their departments.
Similarly, we know that service bureaucracies consistently favor some cli-
ents over others, despite official policies designed to treat people alike. To
understand how and why these organizations sometimes perform contrary
to their own rules and goals, we need to know how the rules are experi-
enced by workers in the organization, what latitude workers have in acting
on their preferences, and what other pressures they experience.
Few callings deserve greater respect than those involving public service.
As citizens we are grateful to those people who teach our children, protect
life and property, manage our natural resources, and help people in need to
access social services. These functions have evolved as hallmarks of inclu-
sive, prosperous societies throughout the world. Some street-level occupa-
tions are highly respected and well-paid. Others, such as some social work-
ers, have a more contested position in society. Some operate in relative
obscurity, whereas others, such as police officers and child protection work-
xiii
Preface
ers, are among those who are often in the news for regrettable develop-
ments. If a child dies while in protective care, or a person is badly treated
while in custody, everyone in those agencies experiences the resulting pub-
lic criticism.
One important way in which street-level bureaucrats experience their
work is in their struggle to make it more consistent with their strong com-
mitments to public service and the high expectations they have for their
chosen careers. People often enter public employment with a commitment
to serving the community. Teachers, social workers, public defenders, and
police officers partly seek out these occupations because they offer socially
useful roles. Yet the very nature of these occupations can prevent recruits
to street-level bureaucracies from coming even close to the ideal concep-
tion of their jobs. Large classes, huge caseloads, and other challenging
workload pressures combine with the contagious distress of clients who
have few resources and multiple problems to defeat their aspirations as ser-
vice workers.
Ideally and by training, street-level bureaucrats should respond to the in-
dividual needs or characteristics of the people they serve or confront. In
practice, they must deal with clients collectively, because work require-
ments prohibit individualized responses. Teachers should respond to the
needs of the individual child; in practice, they must develop techniques to
manage a classroom of children. Police officers should respond to the pre-
senting case; in reality, they must develop techniques to recognize and re-
spond to various types of confrontations, particularly those that threaten
their authority or may pose danger. At best, street-level bureaucrats invent
modes of mass processing that more or less permit them to deal with the
public fairly, appropriately, and thoughtfully. At worst, they give in to favor-
itism, stereotyping, convenience, and routinizing-all of which serve their
own or agency purposes.
Compromises in work practices and attitudes are often rationalized as re-
flecting workers' greater experience on the job, their appreciation of practi-
cal and political realities, or their more realistic assessment of the nature of
the work. But these rationalizations only summarize the prevailing struc-
tural constraints on human service bureaucracies. They are not "true" in an
absolute sense. The teacher who psychologically abandons her commitment
to help every child to read may succumb to a private assessment of the sta-
tus quo in education. But this compromise says nothing about the potential
of individual children to learn or the capacity of the teacher to instruct. This
potential remains intact. It is the system of schooling, the organization of
the schooling bureaucracy, that teaches that children are developmentally
xiv
Dilemmas of the Individual in Public Services
'"slow'" or unmotivated, and that teachers must abandon their early com-
mitments to be an excellent teacher to every child.
In the same way, the criminal justice system allows police recruits to pre-
sume that they can approach with impunity young people hanging out in
certain neighborhoods to see whether they are in possession of guns or
drugs, even if they have no evident cause for suspicion other than the coin-
cidence of age, race, and neighborhood. Young police officers learn that
judges will back them up if the young people claim that the officers planted
evidence or made up their own descriptions of the encounters. Court offi-
cers, judges, prosecutors, and public defenders collaborate in the mass pro-
cessing of a great many new and repeat juvenile offenders each year yet re-
tain the ideal that each may have his or her fair and full "day in court."
Some street-level bureaucrats drop out or bum out relatively early in
their careers. Those who stay on, to be sure, often grow in the jobs and per-
fect treatment and client-processing techniques that provide an acceptable
balance between public aspirations for the work and the coping require-
ments of the job. These adjustments of work habits and attitudes may re-
flect lower expectations for themselves, their clients, and the potential of
public policy. Ultimately, these adjustments permit acceptance of the view
that clients receive the best that can be provided under prevailing circum-
stances.
Street-level bureaucrats often spend their work lives in these corrupted
worlds of service. They believe themselves to be doing the best they can
under adverse circumstances, and they develop techniques to salvage ser-
vice and decision-making values within the limits imposed on them by the
structure of the work. They develop conceptions of their work and of their
clients that narrow the gap between their personal and work limitations and
the service ideal. These work practices and orientations are maintained
even as they contribute to the distortion of the service ideal or put the
worker in the position of manipulating citizens on behalf of the agencies
from which citizens seek help or expect fair treatment.
Should teachers, police officers, or social workers look for other work
rather than participate in practices that seem far from ideal? This would
mean leaving clients to others who have even fewer concerns and less inter-
est in clients than they do. It would mean not only starting over in a new
career, but also abandoning the satisfactory aspects of the work they have
managed to carve out.
Should they stay in their jobs and dedicate themselves to changing client-
processing conditions from within their agencies? This approach is prob-
lematic as well, though it is the career path taken by many who leave direct
XV
Preface
service for management. In their new positions, some will be reformers
striving for change to the limit of their capacity and what the times will
bear. Others will settle for the status quo.
The structure of street-level bureaucracy also confronts clients with di-
lemmas bearing on action. Consumers of public services, for the most part,
cannot choose the public services to which they will be subject. They must
accept the schools, courts, and police forces of their communities. If they
are poor, they must also accept the community's arrangements for health
care, income support, housing subsidies and other benefit programs. In ap-
proaching the institutions that administer these policies, they must strike a
balance between asserting their rights as citizens and conforming to the be-
haviors public agencies seek to place on them as clients. As citizens, they
should seek all to which they are entitled. As bureaucratic subjects, they
must temper their demands in accord with their assessment of the limita-
tions of the public agencies which control benefits and sanctions. Although
it is apparent that exceptions are often made and additional resources often
found, clients also recognize the potential costs of unsuccessfully asserting
their rights.
On matters of the greatest urgency and moment, such as health care, ed-
ucation, justice, housing, and income, clients passively seek support and fair
treatment from public agencies when evidence and experience suggest that
their hopes may go unrewarded. The dilemmas of action may be particu-
larly acute if clients are poor, are immigrants, or are of a different racial or
ethnic background than the public employees with whom they interact.
Should I wait my tum and submit to the procedures of the agency, despite
reservations? I risk being unable to gain attention to my particular needs
and concerns. Should I speak out forcefully and demand my rights? I risk
antagonizing the workers by disrupting office procedures.
It is no small thing to adjust successfully to the rigors of the street-level
workplace. Virtually all jobs involve adjustments to routines of practice,
challenges to keeping a fresh outlook despite repetitive tasks, and compro-
mises between personal needs and vocational requirements. Despite the
many barriers to effective practice described in these pages, street-level bu-
reaucrats frequently manage to find a satisfactory balance between the re-
alities of the job and personal fulfillment. The society is all the better for
their capacity to find a satisfactory balance in their work life.
When I originally wrote this book, I was intent on elaborating on the cop-
ing behaviors of street-level bureaucrats. In doing so I emphasized the gap
between the realities of practice and service ideals. This approach had its
value, to judge from the reception the book has enjoyed over the years. But
xvi
Dilemmas of the Individual in Public Services
it led to neglect of an important reality: vast numbers of people in public
service on a daily basis go to work at rewarding and fulfilling jobs. They
meet their classes, carry out their assignments, and manage their caseloads
without much complaint. Work goes on in these public service organizations
to general satisfaction. Partly this is because workers, clients, and the gen-
eral public have more modest expectations than they might have. But it also
goes on because frontline workers have figured out how to do what they re-
gard as a reasonable job with resources at their disposal.
How does one reconcile a clear-eyed assessment of the dilemmas of
street-level bureaucracy with the reality that many if not most teachers, po-
lice officers, and social workers like what they do and do their jobs relatively
well by community standards?
There are two ways to understand the term "street-level bureaucrat."
One is to equate it with the public services with which citizens typically in-
teract. In this sense, all teachers, police officers, and social workers in pub-
lic agencies are street-level bureaucrats without further qualification. This is
the way the term has commonly come to be used.
Another way-the one I originally intended-was to define street-level
bureaucracy as public service employment of a certain sort, performed un-
der certain conditions. In this second approach, street-level bureaucrats in-
teract with citizens in the course of the job and have discretion in exercising
authority; in addition, they cannot do the job according to ideal conceptions
of the practice because of the limitations of the work structure.
When I first wrote the book, I did not mean to suggest that every front-
line worker experienced stressful working conditions. I know teachers who
with little effort are able to pay attention to every child in their small school.
I know art teachers who experience little stress in leading their classrooms
through exercises. I know National Park rangers whose daily routines re-
quire coping with boredom as much as anything else.
In this second conception of the term, in other words, not every teacher,
police officer, or public social worker experiences the pressures that I stated
street-level bureaucrats face by definition. Frontline workers whose jobs are
relatively free of restrictive structural constraints will still develop routines
in response to their work requirements. But the routines will not be devel-
oped primarily to cope with a difficult work environment. If we adopt the
second perspective, we can see that not every frontline worker experiences
the pressures this book analyzes.
Additionally, although many street-level coping behaviors may widen the
gap between policy as written and policy as performed, other coping behav-
iors reflect acceptable compromises between the goals of enacted policy
xvii
Preface
and the needs of the street-level workers. Not every coping mechanism dis-
tances the worker from the goals of the organization. Indeed, the best work-
ers are the ones who bridge the gap.
Perhaps it is best to imagine a continuum of work experiences ranging
from those that are deeply stressful and the processing of clients is severely
underresourced, to those that provide a reasonable balance between job re-
quirements and successful practice. Workers' places on that continuum
might change over time as they gain experience, as caseloads and assign-
ments vary, or as the workplace itself adopts new approaches or engages
new clienteles. All street-level bureaucrats potentially confront circum-
stances that lead them to coping mechanisms representing departures from
the service ideal. But all frontline workers do not cope with these issues all
the time.
Still another reason that many street-level bureaucrats can successfully
negotiate the gap between the ways they cope with their jobs and public
expectations is that those expectations are undoubtedly considerably lower
than the ideal. Public expectations may replicate on a societal basis the
compromises street-level bureaucrats adopt in coping with their clients per-
son by person.
This can be explained partly by the fact that the work of street-level bu-
reaucrats is mostly hidden from public view, so even attentive citizens do
not necessarily know what is going on agency by agency. Also, to the extent
that general expectations of public services go beyond demands for effi-
ciency and honest administration, they are likely to focus on incremental
improvement. That is, the hopes of the public for improved agency perfor-
mance are likely to focus on marginal changes in client or administrative
outcomes, and are likely to be based on limited indicators. Reformers who
hold out for prospects of radically better services and client outcomes tend
to be dismissed as excessively idealistic.
A final set of dilemmas confronts citizens who are continuously, if implic-
itly, asked to evaluate public services. This occurs in focused forums such as
a referendum on a school budget or a revolt against high property taxes. It
also occurs in diffuse expressions of dissatisfaction with the public sector,
such as Colorado's famous Taxpayers' Bill of Rights (TABOR), which set in
motion a drastic decline in public services until TABOR was suspended by
voters in 2005. Indeed, the many initiatives to limit state and local spending
in recent decades have largely been understood as attacks on the value of
government.
What are the policy alternatives? When all the "fat" has been trimmed
from agency budgets and all the "waste" eliminated, the basic choices re-
xviii
Dilemmas of the Individual in Public Services
main: to further automate, systematize, and regulate the interactions be-
tween government employees and citizens seeking help; to drift with the
current turmoil that favors reduced services and greater standardization in
the name of cost effectiveness and budgetary controls; or to secure or re-
store the importance of human interactions in services that require discre-
tionary intervention or involvement.
How much can human intervention be eliminated from teaching, nurs-
ing, policing, and judging? The fact is that we must have people making
decisions and treating other citizens in the public services. We are not pre-
pared as a society to abandon decisions about people and discretionary in-
tervention to machines and programmed formats. Yet how can one advocate
greater attention to the intervening and discretionary roles of street-level
bureaucrats in the face of the enormous and often well-deserved popular
discontent with the effectiveness and quality of their work?
I try to address these questions in this book. I do not exonerate street-
level bureaucracies, excuse their deficiencies, or urge their support as cur-
rently structured. Rather, I locate the problem of street-level bureaucrats in
the structure of their work, and attempt to identifY conditions that would
better support a reconstituted public sector dedicated to appropriate ser-
vice and respect for clients--one that would be more likely to produce ef-
fective service providers. In developing the street-level bureaucracy frame-
work, I identifY the common elements of occupations as apparently
disparate as, say, police officer and social worker. The analysis of street-level
bureaucracy helps us identifY which features of people-processing are com-
mon, and which are unique, to the different occupational milieux in which
they arise.
Moreover, this essentially comparative approach permits us to raise ques-
tions systematically about apparent differences in various service areas. For
example, recognition that all street-level bureaucracies need to control cli-
ents gives perspective to police officer shows of force and raises questions
about precisely what in the work context of police officers makes client con-
trol so dominant a theme.
Just as one of the most important contributions of the concept of "profes-
sionalism" is to facilitate understanding of the differences between, say,
doctors and nurses, in the same way the concept of street-level bureaucracy
should encourage exploration of important differences in public services as
well as contribute to an understanding of central tendencies that they share.
Street-level bureaucrats are major recipients of public expenditures and
represent a significant portion of public activity at the local level. Citizens
directly experience government through them, and their actions are the
xix
Preface
policies provided by government in important respects. I start by summariz-
ing the importance of street-level bureaucrats in contemporary political life
and explain the sense in which these low-level workers may be understood
to "make" the policies they are otherwise charged with implementing (part
I). Then I treat the common features of street-level work and explore the
implications of these conclusions for client outcomes, organizational con-
trol, and worker satisfaction (part II).
The utility of the street-level bureaucracy approach can be tested only in
efforts to understand whether common features of the framework lead to
common behavioral outcomes. I explore this general question with refer-
ence to street-level tendencies to ration and restrict services, control clients
and the work situation, and develop psychological dispositions that reduce
the dissonance between worker expectations and actual service outcomes
(part III). In the next section, I provide an assessment of the effect of fiscal
crisis on street-level bureaucrats, and a discussion of the potential for re-
form and reconstruction of these critical public functions (part IV).
These latter chapters may be of particular interest to readers of this new
edition for the insight they may provide on developments over the last thirty
years. On the one hand, the implications that reform movements within the
professions might play a restorative role today seem more farfetched than
they did thirty years ago. On the other hand, the choices available to the
society for managing street-level bureaucracies toward greater responsive-
ness and democratic accountability remain reasonably intact. It is also note-
worthy that the theme of fiscal crisis, which dominated discussions of cut-
backs in public services as a result of tax revolts of the late 1970S, are still
with us today. These themes are reviewed and account taken of recent de-
velopments in public services in the final chapter, which was written espe-
cially for this edition.
XX
ACKNOWLEDGMENTS
2010
For more than thirty years I have been privileged to track the development
of street-level bureaucracy as the ideas embodied in this book have rolled
out, been used and evaluated, and deployed in new ways. They serve some
simply to designate classes of actors in the policy process. In other hands
they have been adopted and the framework modified to guide extensive in-
quiries into very specific lines of work. Researchers have organized their re-
search based on the street-level bureaucracy perspective in the study of cus-
toms inspectors in Senegal, employment counselors in Australia, and labor
inspectors in the Dominican Republic. It has been extremely gratifying to
have been able to observe closely this swirl of ideas.
Of all the comments and critiques I have received during this long pe-
riod, two have remained particularly memorable. The first is a set of ac-
counts of many current and former public sector workers who have read the
book, usually on the occasion of having returned to graduate school after a
few years in the field. They say that in reading the book they recognize
themselves and their struggles at work. They report that the book helped
them feel better about the way they adapted to life in the organization. The
difficulties they were having at work, they now understood, were not neces-
sarily attributable to their personal failings, but instead at least in part were
the result of the structure of their jobs. I particularly appreciate these com-
ments because a first step in empowerment of the individual is recognizing
the systemic basis of one's condition or circumstances.
As to the second, a few years after the book was published I agreed to be
interviewed by telephone from my office at MIT by students in North Da-
kota who were studying to be social workers. One student thought the book
was very persuasive but, she said essentially, "You paint such a grim pic-
ture-after reading your book I don't know whether I want to go into the
field!" I was taken aback, but she was right. Whatever the value of the book
xxi
Acknowledgments
for researchers and policy analysts, I understood that for people considering
careers in public service the book might well be discouraging.
I have tried to address this concern here, in the new preface to the book,
but the topic deserves much greater attention. Literally millions of people
choose to go into the public sector because of the rewards and challenges of
working with and for other people. They deserve much more public ap-
proval and respect (and usually higher pay) than they generally receive. In
support of these choices, it would be good to know, from systematic re-
search conducted from a street-level bureaucracy perspective, how police
officers, teachers, and social workers find that satisfactory balance between
the expectations of the job and what they are able to accomplish.
I thank the Russell Sage Foundation for keeping the book in print over
all these years, and inviting me to revise the preface and write a new chap-
ter for this edition. I am very grateful to Robert Behn, Evelyn Brodkin, Car-
olyn Hill, Deborah Stone, Steven Rathgeb Smith, and Soeren Winter,
friends and distinguished scholars, who offered comments on a draft of the
new chapter on painfully short notice. I owe special thanks and recognition
to my colleagues at Demos, who provide a remarkable home for researchers
and activists committed to striving for an inclusive democracy and shared
prosperity, supported by an effective and responsive public sector.
1g8o
xxii
Acknowledgments
erty at the University of Wisconsin, and the Department of Political Sci-
ence and the Graduate School of Public Affairs of the University of
Washington. Graduate students at the University of Washington, students
at the College of Public and Community Service of the University of
Massachusetts, Boston, as well as graduate students at M.I.T. have taught
me a great deal about street-level bureaucracy over this period.
Many friends and colleagues have contributed to this work in conver-
sation, in their writings, or through expressions of personal interest and
support. I am particularly grateful in various measures to Robert Alford,
Gary Bellow, Murray Edelman, Willis Hawley, Ira Katznelson, Jeanne Ket-
tleson, Margaret Levi, Hannah Lipsky, David J. Olson, Jeffrey Pressman,
Martin Rein, Charles Sabel, and Aaron Wildavsky. My debt to many
other writers who have written usefully on public services that function
as street-level bureaucracy is recognized in the notes. I am pleased to
acknowledge my double debt to Martha Wagner Weinberg, first as tire-
less and inventive research assistant, and second (many years later) as val-
ued colleague at M .I. T.
Many of the ideas that comprise this book were developed in collabora-
tion with Carl Hosticka, Jeffrey Prottas, and Richard Weatherley when
they served as research assistants under the Russell Sage grant. I am proud
of that association and deeply appreciate their many insights and contri-
butions to our common enterprise.
This book has been greatly influenced by Suzanne Lipsky. Among her
many contributions has been her recognition and analysis of the potential
of people to sustain and recover their humanity despite contributing
to or being subjects of oppressive social systems.
xxiii
PART I
INTRODUCTION
CHAPTER 1
3
INTRODUCTION
4
The Critical Role of Street-Level Bureaucrats
enormous number of public workers share these job characteristics. They
comprise a great portion of all public employees working in domestic affairs.
State and local governments employ approximately 37 million people in
local schools, more than 500,000 people in police operations, and over
300,000 people in public welfare. Public school employees represent more
than half of all workers employed in local governments. Instructional jobs
represent about two-thirds of the educational personnel, and many of the
rest are former teachers engaged in administration, or social workers, psy-
chologists, and librarians who provide direct services in the schools. Of the
3.2 million local government public employees not engaged in education,
approximately 14 percent work as police officers. One of every sixteen jobs
in state and local government outside of education is held by a public welfare
worker. 2 In this and other areas the majority of jobs are held by people with
responsibility for involvement with citizens.
Other street-level bureaucrats comprise an important part of the re-
mainder of local government personnel rolls. Although the U.S. Census
Bureau does not provide breakdowns of other job classifications suitable for
our purposes, we can assume that many of the 1.1 million health workers, 3
most of the 5,000 public service lawyers, 4 many of the employees of the
various court systems, and other public employees also perform as street-
level bureaucrats. Some of the nation's larger cities employ a staggering
number of street-level bureaucrats. For example, the 26,68o school teachers
in Chicago are more numerous than the populations of many of the Chicago
suburbs. 5
Another measure of the significance of street-level bureaucrats in public
sector employment'is the amount of public funds allocated to pay them. Of
all local government salaries, more than half went to public education in
1973. Almost 8o percent of these monies was used to pay instructional per-
sonnel. Police salaries comprised approximately one-sixth of local public
salaries not assigned to education. 6
Much of the growth in public employment in the past 25 years has oc-
curred in the ranks of street-level bureaucrats. From 1955 to 1975 govern-
ment employment more than doubled, largely because the baby boom of the
postwar years and the growing number of elderly, dependent citizens in-
creased state and local activity in education, health, and public welfare. 7
Street-level bureaucracies are labor-intensive in the extreme. Their busi-
ness is providing service through people, and the operating costs of such
agencies reflect their dependence upon salaried workers. Thus most of what-
ever is spent by government on education, police, or other social services
(aside, of course, from income maintenance, or in the case of jails and
5
INTRODUCTION
6
The Critical Role of Street-Level Bureaucrats
reformers move to separate service provision from decisions about support
payments, or they design a negative income tax system that would eliminate
social workers in allocating welfare. Problems of backlog in the courts are
met with proposals to increase the number of judges. Recognition that early-
childhood development largely establishes the potential for later achieve-
ment results in the development of new programs (such as Head Start) in
and out of established institutions, to provide enriched early-childhood
experiences.
In the 1g6os and early 1970s the modal governmental response to social
problems was to commission a corps of street-level bureaucrats to attend to
them. Are poor people deprived of equal access to the courts? Provide them
with lawyers. Equal access to health care? Establish neighborhood clinics.
Educational opportunity? Develop preschool enrichment programs. It is far
easier and less disruptive to develop employment for street-level bureau-
crats than to reduce income inequalities.
In recent years public employees have benefitted considerably from the
growth of public spending on street-leve! bureaucracies. 12 Salaries have
increased from inadequate to respectable and even desirable. Meanwhile,
public employees, with street-level bureaucrats in the lead, have secured
unprecedented control over their work environments through the develop-
ment of unions and union-like associations. 13 For example, teachers and
other instructional personnel have often been able to maintain their posi-
tions and even increase in number, although schools are more frequently
under attack for their cost to taxpayers. The ratio of instructional personnel
in schools has continued to rise despite the decline in the number of school-
age children. 14 This development supplements general public support for
the view that some street-level bureaucrats, such as teachers and police of-
ficers, are necessary for a healthy society. us
The fiscal crisis that has affected many cities, notably New York and more
recently Cleveland and Newark, has provided an opportunity to assess the
capacity of public service workers to hold onto their jobs in the face of enor-
mous pressures. Since so much of municipal budgets consists of inflexible,
mandated costs-for debt service., pension plans and other personnel bene-
fits, contractually obligated salary increases, capital expenditure commit-
ments, energy purchases, and so on-the place to find "fat" to eliminate
from municipal budgets is in the service sector, where most expenditures
tend to be for salaries. While many public employees have been fired during
this crisis period, it is significant that public service workers often have been
able to lobby, bargain, and cajole to minimize this attrition. 18 They are sup-
ported in their claims by a public fearful of a reduced police force on the
7
INTRODUCTION
street and resentful of dirtier streets resulting from fewer garbage pickups.
They are supported by families whose children will receive less instruction
from fewer specialists than in the past if teachers are fired. And it does not
hurt their arguments that many public employees and their relatives vote in
the city considering force reductions. 17
The growth of the service sector represents the furthest reaches of the
welfare state. The service sector penetrates every area of human needs as
they are recognized and defined, and it grows within each recognized area.
This is not to say that the need is met, but only that the service state
breaches the barriers between public responsibility and private affairs.
The fiscal crisis of the cities focuses on the service sector, fundamentally
challenging the priorities of the service state under current perceptions of
scarcity. Liberals have now joined fiscal conservatives in challenging service
provision. They do not do so directly, by questioning whether public ser-
vices and responsibilities developed in this century are appropriate. Instead,
they do it backhandedly, arguing that the accretion of public employees and
their apparently irreversible demands upon revenues threaten the au-
tonomy, flexibility, and prosperity of the political order. Debates over the
proper scope of services face the threat of being overwhelmed by challenges
to the entire social service structure as seen from the perspective of unbal-
anced public budgets.
8
The Critical Role of Street-Lev el Bureaucra ts
of some citizens to government al goods and services at the expense of gen-
eral taxpayers and those whose claims are denied. By increasing or decreas-
ing benefits availability to low-income recipient populations they implicitly
regulate the degree of redistributio n that will be paid for by more affluent
sectors.
In another sense, in delivering policy street-level bureaucrats make deci-
sions about people that affect their life chances. To designate or treat some-
one as a welfare recipient, a juvenile delinquent, or a high achiever affects
the relationships of others to that person and also affects the person's self-
evaluation. Thus begins (or continues) the social process that we infer
accounts for so many self-fulfilling prophecies. The child judged to be ajuve-
nile delinquent develops such a self-image and is grouped with other "delin-
quents," increasing the chances that he or she will adopt the behavior
thought to have been incipient in the first place. Children thought by their
teacher to be richly endowed in learning ability learn more than peers of
equal intelligence who were not thought to be superior. 18 Welfare recipients
find or accept housing inferior to those with equal disposable incomes who
are not recipients.19
A defining facet of the working environmen t of street-level bureaucrats is
that they must deal with clients' personal reactions to their decisions, how-
ever they cope with their implications. To say that people's self-evaluation is
affected by the actions of street-level bureaucrats is to say that people are
reactive to the policy. This is not exclusively confined to subconscious pro-
cesses. Clients of street-level bureaucracie s respond angrily to real or per-
ceived injustices, develop strategies to ingratiate themselves with workers,
act grateful and elated or sullen and passive in reaction to street-level bu-
reaucrats' decisions. It is one thing to be treated neglectfully and routinely
by the telephone company, the motor vehicle bureau, or other government
agencies whose agents know nothing of the personal circumstanc es sur-
rounding a claim or request. It is quite another thing to be shuffied, catego-
rized, and treated "bureaucrati caily," (in the pejorative sense), by someone
to whom one is directly talking and from whom one expects at least an open
and sympathetic hearing. In short, the reality of the work of street-level bu-
reaucrats could hardly be farther from the bureaucratic ideal of impersonal
detachment in decision making. 20 On the contrary, in street-level bureau-
cracies the objects of critical decisions-p eople-actua lly change as a result
of the decisions.
Street-level bureaucrats are also the focus of citizen reactions because
their discretion opens up the possibility that they will respond favorably on
behalf of people. Their general and diffuse obligation to the "public interest"
9
INTRODUCT ION
pennits hope to flourish that the individual worker will adopt a benign or fa-
vorable orientation toward the client. Thus, in a world of large and imper-
sonal agencies that apparently hold the keys to important benefits, sanctions,
and opportunities, the ambiguity of work definitions sustains hope for a
friend in court.
This discussion helps explain continued controversy over street-level bu-
reaucracies at the level of individual service provision. At the same time, the
peculiar nature of government service delivery through street-level bureau-
crats helps explain why street-level bureaucracies are apparently the pri-
mary focus of community conflict in the current period, and why they are
likely to remain the focus of such conflict in the foreseeable future. It is no
accident that the most heated community conflicts since 1964 have focused
on schools and police departments, and on the responsiveness of health and
welfare agencies and institutions. 21 These are the sites of the provision of
public benefits and sanctions. They are the locus of individual decisions
about and treatment of citizens, and thus are primary targets of protest. As
Frances Fox Piven and Richard Cloward explain:
... people experience deprivation and oppression within a concrete setting, not as
the end product oflarge and abstract processes, and it is the concrete experience that
molds their discontent into specific grievances against specific targets .... People on
relief [for example] experience the shabby waiting rooms, the overseer or the case-
worker, and the dole. They do not experience American social welfare policy .... In
other words, it is the daily experience of people that shapes their grievances, es-
tablishes the measure of their demands, and points out the targets of their anger. 22
10
The Critical Role of Street-L evel Bureaucr ats
directed at communit y institutions, their focus in protesting the actions of
street-leve l bureaucrac ies may be attributed to the familiarity of the agency,
its critical role in communit y welfare, and a perception at some level that
these institutions are not sufficiently accountable to the people they serve.
Finally, street-leve l bureaucra ts play a critical role in regulating the de-
gree of contempor ary conflict by virtue of their role as agents of social con-
trol. Citizens who receive public benefits interact with public agents who
require certain behaviors of them. They must anticipate the requireme nts of
these public agents and claimants must tailor their actions and develop "suit-
able" attitudes both toward the services they receive and toward the street-
level bureaucra ts themselve s. Teachers convey and enforce expectations of
proper attitudes toward schooling, self, and efficacy in other interactions.
Policemen convey expectations about public behavior and authority. Social
workers convey expectations about public benefits and the status of
recipients.
The social control function of street-leve l bureaucra ts requires comment
in a discussion of the place of public service workers in the larger society.
The public service sector plays a critical part in softening the impact of the
economic system on those who are not its primary beneficiaries and inducing
people to accept the neglect or inadequacy of primary economic and social
institutions. Police, courts, and prisons obviously play such a role in proces-
sing the junkies, petty thieves, muggers, and others whose behavior toward
society is associated with their economic position. It is a role equally played
by schools in socializing the population to the economic order and the likely
opportunit ies for different strata of the population. Public support and em-
ployment programs expand to ameliorate the impact of unemploy ment or
reduce the incidence of discontent ; they contract when employme nt oppor-
tunities improve. Moreover, they are designed and implemen ted to convey
the message that welfare status is to be avoided and that work, however
poorly rewarded, is preferable to public assistance. One can also see the two
edges of public policy in the "war on poverty" where the public benefits of
social service and communit y action invested neighborh ood institutions with
benefits for which potential dissidents could compete and ordinary citizens
could develop dependenc y. 24
What to some are the highest reaches of the welfare state are to others the
furthest extension of social control. Street-lev elbureauc rats are partly the
focus of controvers y because they play this dual role. Welfare reform
founders on disagreem ents over whether to eliminate close scrutiny of wel-
fare applications in order to reduce administra tive costs and harassmen t of
recipients, or to increase the scrutiny in the name of controlling abuses and
11
INTROD UCTION
12
CHAPTER 2
Street-Level Bureaucrats
as Policy Makers
Discretion
13
INTRODUCTION
the one hand, and impartiality and rigid rule-application on the other hand
presents a dialectic of public service reform. Reformers attempt to limit
worker discretion at one time, and increase it at another. In order to make
ambulance dispatch by untrained personnel more efficient, health planners
work to rationalize emergency dispatch procedures by developing a pro-
grammed format to aid in identifying health emergencies. 10 Meanwhile,
other health planners seek to replace untrained admission clerks with health
professionals in order to insure greater sensitivity to the health problems of
prospective patients. 11 Programmed learning materials are introduced to
release teachers for more intensive work with some student&, while all stu-
dents benefit from these motivational and self-paced features. Later this in-
novation is found wanting because it eliminated teachers' feedback to stu-
dents and encouraged regimentation rather than individualized learning. 12
To the extent that tasks remain complex and human intervention is consid-
ered necessary for effective service, discretion will remain characteristic of
many public service jobs.
Most analysts take for granted that the work of lower-level participants will
more or less conform to what is expected of them. Organizational theorists
recognize that there will always be some slippage between orders and the
carrying out of orders, but this slippage is usually attributed to poor com-
munication or workers' residual, and not terribly important, disagreement
with organizational goals. In any event, such difficulties are usually consid-
ered unimportant enough that organizations can overcome them.
This observation is partly derived from the recognition that lower level
workers' behavior in organizations, including public agencies, appears to be
cooperative. Workers for the most part accept the legitimacy of the formal
structure of authority, and they are not in a position to dissent successfully.
But what if workers do not share the objectives of their superiors? Lower-
level participants in organizations often do not share the perspectives and
preferences of their superiors and hence in some respects cannot be thought
to be working toward stated agency goals. At least this is the case when
workers are not recruited with an affinity for the organization's goals; work-
ers do not consider orders from "above" legitimate; or the incentives avail-
able to supervisors are matched by countermeasures available to lower-level
16
Street-Level Bureaucrats as Policy Makers
participants. One can expect a distinct degree of noncompliance if lower-
level workers' interests differ from the interests of those at higher levels, and
the incentives and sanctions available to higher levels are not sufficient to
prevail. 13
Sometimes different levels of organizations are more appropriately cou-
ceived as intrinsically in conflict with each other rather than mutually re-
sponsive and supportive. 14 At times it is more useful to view lower-level
workers as having distinctly different interests and the resources to pursue
those interests. Here discrepancies between policy declarations and actual
policy would be expected and predictable. And explanations for the discrep-
ancies would be searched for not in the breakdown or inadequacy of the
compliance system but in the structure of the work situation from which
workers' "antagonistic" interests arise.
In such organizations policy may be carried out consistent with the inter-
ests of higher levels, but this should be understood as resulting from the mu-
tual adjustment of antagonistic perspectives as well as the result of shared in-
terests. In such cases the latent conflict in the interests of different levels is
suppressed or is a matter ofindifference to one or both parties. This approach
takes as problematic the mutuality of interests, and it searches instead for
the mechanisms by which essentially antagonistic or divergent interests are
adj usted. 15
Some of the ways lower-level workers can withhold cooperation within
their organizations include such personal strategies as not working (excessive
absenteeism, quitting), aggression toward the organization (stealing, cheat-
ing, deliberate wasting), and negative attitudes with implications for work
(alienation, apathy). 16 Workers may take advantage of collective resources to
act nonco.Jperatively by forming trade unioos or by exercising rights under
collective bargaining agreements or civil service regulations. These collec-
tive strategies for noncooperation contribute to workers' willingness to dis-
play lack of motivation and to perform at only minimal levels. 17
These forms of noncooperation injure organizations' abilities to achieve
their objectives because workers perform at less than full capacity. The man-
agement challenge perceived to be at the heart of the problem is how to
make workers' needs for personal, material, or psychological gratification
mesh with the organization's needs. Thus the management problem regard-
ing worker absenteeism becomes how to improve job satisfaction while re-
taining productivity.
However, there is another class of conflicts between lower-level workers
and the organizations that arise not from the personal needs of the workers
alone but also from their positions within their organizations. The role of
INTRODUCTION
street-level bureaucrats, like other roles, may be conceived as a set of ex-
pected interests 18 as well as expected behaviors. Street-level bureaucrats
may be shown to have distinctly different interests from the interests of
others in the agencies for which they work. Moreover, certain features of
their role make it possible for them to make these differences manifest. Dif-
ferences in interests, and the possibility of surfacing the differences, permit
analysis of the structural position of street-level bureaucrats from a conflict
perspective. 19
In the following brief discussion, the nature of the conflict between the
objectives and orientations of street-level bureaucrats and those in higher
authority roles is considered first. Then the capacity of street-level bureau-
crats' ability to resist organizational directives is treated. 20
programs to provide first offenders with counseling, job training, and place-
ment assistance, and alcoholics, reckless drivers, and drug offenders with ap-
propriate counseling. In addition, judges have the services of psychiatrists,
social workers, probation officers, and others who might be able to provide
treatment as an alternative to imprisonment. These developments have
been conceived by humanitarian reformers who believe, along with many
judges, that prisons create more criminals than they deter by exposing peo-
ple to experienced crooks, and by pragmatists, who recognize that the courts
have become revolving doors of repeat appearances without deterrent effect.
It is conspicuous to court observers that these programs take a heavy bur-
den off the judge. The judge is now able to make what appears to be a con-
structive decision rather than simply to choose between the unattractive al-
ternatives of sending a person to jail or releasing the putative offender
without penalty. Indirect evidence that these programs fill critical institu-
tional needs is suggested by the Boston pretrial diversion programs. These
programs were utilized beyond their capacity by judges, sometimes without
regard for the extremely important initial interview or the relatively strin-
gent eligibility requirements the programs sought to impose in order to max-
imize effectiveness. Dependent upon judges for referrals and, indeed, for
their programs' existence, administrators found it difficult to refuse judges
who referred too many clients, or inappropriate clients, to them. 23
The Veterans Administration hospital system is a fascinating bureaucracy
because it employs doctors, the preeminent professionals, in highly rule-
bound organizations. The country's system of socialized medicine for the in-
digent veteran has developed an extremely complex series of rules because
of simultaneous congressional concern to provide veterans with hospital ser-
vices, maintain strict cost accounting, and (particularly in the past) not com-
pete with private medical practice. In large part because the VA system was
to provide hospital care, leaving to private physicians the business of office
consultations, the VA hospitals were prohibited from treating patients on an
outpatient basis. This conflicted with the doctors' prerogative to prescribe a
level of treatment appropriate to the problem presented. There are many
patients who need treatment, but not hospitalization, and to hospitalize such
people would be to deny others the hospital space.
However, there was an allowable exception to the rule limiting services to
hospitalized veterans. Under the "Pre-Bedcare" category (PBC) veterans
who required health services prior to their anticipated admission (for ex-
amp!e, blood tests prior to surgery) could be treated. Despite various
requirements intended to limit PBC treatment to those whose admission
was clearly expected, actual admission to the hospitals from PBC lists was
20
Street-Level Bureaucrats as Policy Makers
traditionally very low. It seems that doctors, chafing under the restriction
that they could not treat patients according to their best estimate of need,
were treating patients as outpatients under the fiction that they were ex
pected to be admitted. This supposition is supported by observing what hap-
pened when the VA introduced an Ambulatory Health Care (AHC) policy,
permitting outpatient treatment for the first time. PBC admissions plum-
meted, while ABC admissions rose steadily. In one hospital AHC received
148 patients in the first five months of the program, while PBC dropped
from 122 to 73 patients during that period. Fifty-one of these patients trans-
ferred directly to the AHC program. 24 In retrospect it seems that doctors
were able to utilize existing bureaucratic structures to impose their views of
proper treatment on the organization, despite organizational efforts to cir-
cumscribe their discretion.
Street-level bureaucrats will also use existing regulations and administra~
tive provision to circumvent reforms which limit their discretion. In Decem-
ber 1968, in response to pressure from the Department of Housing and
Urban Development, the Boston Housing Authority (BHA) adopted new
tenant-selection guidelines designed to insure housing project racial integra-
tion. The plan utilized what was known as the "1-2-3 rule." To eliminate per-
sonnel discretion in assignments the 1-2-3 rule provided that prospective
tenants would be offered places only in the housing projects with the three
highest vacancy ratios. If these offers were refused, the application would be
returned to the bottom of the waiting list.
The BHA integration plan did not work. Many housing authority em-
ployees objected to assigning people to projects in which they did not want
to live. They were particularly concerned for their traditionally favored cli-
entele, the elderly, poor whites who populated the "better" BHA projects.
Among the reasons the reform did not work were that housing authority per-
sonnel were so inundated with work that proper adminstrative controls were
not feasible, and in the chaos of processing applications, those who wished to
favor some prospective tenants over others were able to do so. Housing of-
ficials took advantage of provisions for exceptions to the 1-2-3 rule, interpret-
ing reasonable provisions for flexibility in extremely liberal ways when they
wanted to. They volunteered information to favored prospective tenants
concerning how to have their applications treated as emergencies or other
high-priority categories, while routinely processing the applications of oth-
ers. Applications were frequently lost or misplaced so that workers could
favor tenants simply by locating their files and acting on them, while other
files remained unavailable for processing. Meanwhile, public housing man-
agers contributed to the sustained biases of the agency by failing to report
21
INTRODU CTION
vacancies to the central office when they occurred, not informing prospec-
tive tenants when units were available, or showing tenants they wished to
discourage only unattractive or unsafe units, although others in the project
were available. Thus the press of work combined with workers' desires to
continue to serve particular clients restored the discretionary powers the
new rules were designed to eliminate. 25
The power of street-level bureaucrats to thwart reforms by manipulating
legal loopholes i!F suggested in a strategy paper prepared by New York City
officials to reduce welfare rolls. It indicates managerial awareness of the
limits of agency reform without worker cooperation. The paper rejects the
possibility of shortening intake hours too drastically, in part because of the
chance that "welfare workers would go out of the centers and take applica-
tions from people at home, on the street, etc., since application may be
made [under existing Federal law] at agency office, in own home, by tele-
phone, by mail, or in any suitable place. " 26
A dramatic example of the instinct to retain discretionary options is pro-
vided by the response to the New York State laws sponsored by then Gover-
nor Nelson Rockefeller to impose mandatory, severe jail sentences for drug
dealers, while providing relatively minor penalties for those caught with
small amounts of drugs. The rationale for the law was to provide a strong de-
terrent to the drug trafficker. Predictably, some people arrested were not
the rapacious drug dealers whom the law was designed to deter. This was the
case, for example, with drug addicts on methadone who occasionally sold a
dosage. The law presented a dilemma for court personnel, who believed that
the mandatory, minimum sentence (life imprisonment for the methadone
hustlers) was too severe for the offense. In these cases New York District At-
torney Richard Kuh began to charge the alleged offenders not with the
crimes which they had committed, but rather with crimes for which the
punishments were compatible with what he conceived to be the severity of
the offense. In this way, the district attorney attempted to provide the dis-
cretion demanded by a just court system in the face of legislation designed
precisely to eliminate this discretion. 27
Still another source of sustained differences in the interests of street-level
bureaucrats and managers is their continuous interaction with clients and
the varying degrees of complexity in this interaction. Modem bureaucracies
gain legitimacy by (often rhetorical) commitments to standards of fairness
and equity. But street-level bureaucrats are constantly confronted with the
apparent unfairness of treating people alike (just as they recognize the obvi-
ous inequities of unequal treatment). Since individuals are so much more
than their bureaucratically relevant characteristics -age, sex, place of resi-
22
Street-Level Bureaucrats as Policy Makers
dence, income level, etc.-a failure to recognize these differences some-
times seems unfair in itself. The public housing manager believes that there
are degrees of need for public housing not circumscribed by the categories of
eligibility. The teacher recognizes that all children deserve her or his atten-
tion but thinks some require more attention than others.
Not only are the standards of fairness insufficient to dictate levels of con-
cern, but also street-level bureaucrats, like everyone else, have personal
standards of whether or not someone is deserving. The public housing man-
ager may be more sympathetic to the elderly than to other eligibles. Hous-
ing inspectors may be sympathetic to the plight of the landlord although
nothing in the formal structure of the agency encourages such a bias. 28
Under some circumstances it may be appropriate to apply standards of
service with respect to personal characteristics. Street-level bureaucrats
enjoy considerable discretion in part because society does not want compu-
terized public service and rigid application of standards at the expense of
responsiveness to the individual situation. The New York district attorney
was praised edits>rially for contriving to circumvent the mandatory sentenc-
ing requirement. The VA doctors were acting on behalf of patients and in op-
position to limitations on practice when they assigned pre-bedcare status to
patients who did not require hospitalization. Clearly discretion provides op-
portunity to intervene on behalf of clients as well as to discriminate among
them. Yet at best bureaucracies are highly ambivalent about personalistic
service delivery. At the least it is an enduring source of conflict between the
objectives of managers and workers.
2J
INTRODUCTION
25
PART II
CONDITIONS OF
WORK
Introduction
1. Resources are chronically inadequate relative to the tasks workers are asked to
perform.
z. The demand for services tends to increase to meet the supply.
3 Goal expectations for the agencies in which they work tend to be ambiguous,
vague, or conflicting.
CONDITIONS OF WORK
With the possible exception of the last item, to some degree these charac-
teristics follow from the definition of street-level bureaucrats. However,
they are worth elaborating because they call into play various behaviors in
which we are interested.
Note that these conditions of work may not always prevail. For example, it
may be that the welfare department in a small city with a stable, homoge-
neous, white. population is large enough to provide a relatively full range of
social services to recipients and perform with relatively clear objectives
derived from the relatively homogeneous political culture of the city. Em-
ployment and family patterns may be such that there is not a great deal of
movement on or off the welfare roles. In such a situation one might expect
social workers in that office to behave quite differently from the social work-
ers in a large, more heterogeneous central city.
The analysis presented here depends upon the presence of the aforemen-
tioned working conditions. If for some reason these characteristics are not
present, the analysis is less likely to be appropriate, although it is instructive
to understand why this is the case. If a legal services office encouraged its
staff to take only four or five cases at a time in order to maximize the quality
of preparation in each case, the lawyers would behave differently than if they
worked in an office with much higher demands.
Part II provides a detailed discussion of these work conditions and ana-
lyzes their contribution to the problem of providing discretionary social ser-
vices. Part III builds on this analysis to explore the mechanisms street-
level bureaucrats develop to cope with the difficulties and ambiguities inher-
ent in their work.
CHAPTER 3
29
CONDITIONS OF WORK
vices lawyers, while responsible for perhaps 8o to 100 clients whose cases
they currently represent, may typically be working actively on only a dozen
or so cases. 3 Social workers are unable to make required home visits in
public welfare work and are so inundated with paperwork that they are
never without a backlog. 4
Lower-court judges are typically inundated with cases, often causing
delays of several months in providing defendants their day in court. Some-
how judges must accommodate pressures for speedier trials within the con-
straints offairness and equity. 5 In lower courts that hear misdemeanors the
volume is likely to be even greater than in felony courts, although the stan-
dards of justice are presumably the same.
For teachers, overcrowded classrooms (with meager supplies) mean that
they are unable to give the kind of personal attention good teaching
requires. High student-teacher ratios also mean that teachers must attend to
maintaining order and have less attention for learning activities.
For police officers, the obvious resource constraint is one of time-time to
collect information, time to act. A policeman breaking up a fight in a bar does
not have time to determine the initiating party and so must make a double
arrest. 6 Time to develop even a minimum degree of certainty appears to be
unavailable to officers who shoot innocent people because they feared the
consequences of what they say was the civilian's intention to reach for a
weapon. The problem of having to make a quick decision in life-threatening
circumstances is a major foundation of occupational conservatism in police
practice. Police reformers are unable to deal adequately with the heartfelt
claims of policemen that they require maximum discretion in order to pro-
tect themselves. This is the case even though their quick reactions may
evoke the very behavior they seek to defend themselves against, as in the
case of the citizen whose outrage results from the officer's apparent overreac-
tion to a situation.
Clearly, high case loads affect time for decision making. One observer of a
court that heard both felonies and misdemeanors observed that 72 percent of
the cases were handled in one minute or less. 7 Even when physical threat is
not particularly present, street-level bureaucrats must make quick decisions
because of the social reality that they are in the presence of clients who will
interpret indecision as incompetence or lack of authority, with consequences
for subsequent client interactions.
There are other organizational factors that affect the work of street-level
bureaucrats. An emphasis on housekeeping chores, such as filling out forms
or drawing up lesson plans, affects the amount of time available to clients. A
social worker who spends 6o percent of his or her time doing paperwork has
30
The Problem of Resources
correspondingly less time for client interaction. 8 Support services (secre-
taries, clerks, investigators, receptionists) can affect the extent to which
street-level bureaucrats have time for clients. However, efforts to free
street-level bureaucrats of routine tasks so that they may attend to more im-
portant aspects of their work do not necessarily reduce the tensions as-
sociated with that work or improve the quality of interactions between work-
ers and clients. 9 The reasons for this are elaborated in the next section.
Street-level bureaucrats may also lack personal resources in conducting
their work. They may be undertrained or inexperienced. Police rookies, for
example, must undergo a long period of informal apprenticeship before they
are fully accepted and trusted by more experienced veterans. 10 The recent
law school graduate who enters the world of legal services is normally un-
trained both in interaction with clients and aspects of procedure that bear on
the kinds of problems clients are likely to face.
Street-level bureaucrats often experience their jobs in terms of inadequate
personal resources, even when part of that inadequacy is attributable to the
nature of the job rather than rooted in some personal failure. Some jobs just
cannot be done properly, given the ambiguity of goals and the technology of
particular social services. Determining whether or not a job is "proper" is
subject to uncertainties and implicit negotiation with others. (See chapter
10.) For example, there is no consensus on what techniques or approaches
substantially reduce rates of criminal recidivism, although social workers,
psychiatrists, and prison officials are supposed to be responsible for rehabilitat-
ing offenders. 11
From management's perspective street-level bureaucrats are resource
units to be applied to a task. But because of the nature of their tasks, workers
experience their work situations as individuals. One important respect in
which street-level bureaucrats respond as individuals is to the stress under
which they often work.
This is most obvious in the critical case of policemen, whose behavior
often can be explained only by their felt need to avoid danger. They con-
stantly work under the threat of violence that may come from any direction
at any time. Because the threat is unpredictable it exists constantly, although
the actual likelihood of threats materializing is quite low. 12 Police not only
guard against threat but they also characteristically conduct themselves so as
to determine whether threatening situations are likely to exist. They tend to
be lenient with offenders whose attitude and demeanor denote penitence
but harsh and punitive to those offenders who show signs of disrespect. 13
Indeed policemen often appear to test the extent to which an offender re-
spects police authority in order to determine whether he or she is likely to
31
CONDITIONS OF WORK
As a police officer, ... I found myself forced to make the most critical choices in a
time frame of seconds rather than days: to shoot or not to shoot, to arrest or not to ar-
rest, to give chase or to let g<r-always with a nagging certainty that others, those
with great amounts of time in which to analyze and think, stood ready to judge and
condemn me for whatever action I might take or fail to take. o
20
o 0
32
The Problem of Resources
These points suggest that the salience of solutions to problems of resource
inadequacy varies not only with the demands on service and the resources
available, but also with the importance to an individual of deriving a satisfac-
tory solution to these problems.
Most executives profess that their organizations do not have sufficient re-
sources or at least are hampered by resource constraints. Local governments
that tum back federal revenue-sharing funds are the exception to the general
rule that organizations perceive themselves as requiring and able to utilize
additional resources. Yet street-level bureaucracies, with certain other gov-
ernment agencies, chronically experience resource constraints. These agen-
cies are virtually never adequately provided for, and perhaps cannot be ade-
quately provided for. Why is this so?
A distinct characteristic of the work setting of street-level bureaucrats is
that the demand for services tends to increase to meet the supply. If addi-
tional services are made available, demand will increase to consume them. If
more resources are made available, pressures for additional services utilizing
those resources will be forthcoming.
The analogy to the development of traffic patterns on the Long Island
Expressway is compelling. In the name of relieving congestion during rush
hours on this infamous highway, traffic engineers added additional lanes.
But every additional lane, while marginally decreasing driving time to New
York City, induced more people to use the road. This additional traffic re-
stored the traffic jam that the new lanes had been designed to correct.
Utilization increased to meet the supply of road space until commuting
time reached the previous level. A new equilibrium was restored with the
same degree of congestion during rush hours, although with a higher volume
of traffic.
It has often been observed that utilization increases when public services
are expanded. Hospital emergency rooms become inundated because they
provide free medical care or easy access to care at a time when other health
resources, such as family physicians, are less and less available. 21 Mental
health centers discover that they may succeed only too well and evoke such a
33
CONDITIONS OF WORK
large response from citizens that they have to cut back service. 22 Additional
judges do not necessarily decrease court delays because with more judges
courts may be more tolerant of lawyers' delaying tactics, thus restoring the
lengthy time it takes to bring cases to trial. 23 Neighborhood multiservice-
center workers discover they have to abandon plans for systematic client
recruitment because the services of the new program are in such demand
that new cases prove overwhelming and case finding proves superfluous. 24
One dimension of service demand is quantitative. Public expectations of
and demands for certain public services increase over time. In the case of the
police, for example, society expects them to intervene in many more social
conflicts-interracial violence, black assaults on blacks, family disputes, ju-
venile justice-than was the case perhaps forty years ago. 25 Public expecta-
tions of personal health care services have become much greater over time,
spurred, to be sure, by developments in medical technology and health
insurance.
There is a great reservoir of demand for public services generally. In the
area of health, as in virtually every other area in which street-level bureau-
crats operate, there is no imaginable limit to the amount of health care the
population would seek and absorb if it were truly a "free good," available
without significant explicit or implicit costs. For some time the greater use of
physicians by middle-class people was thought by some to be simply a func-
tion of their greater willingness to use doctors, despite the fact that poor
health is inversely related to income. Studies have shown, however, that
when health care is made available at low cost there is little difference be-
tween people of different incomes in their search for medical attention. 26 It
is beyond the present capacity of health training institutions to train the
number of health professionals it would take to equalize the proportion of
health professionals throughout the nation.
More precise and narrowly focused examples of the latent demand for
street-level bureaucrats' services are available. At the end of the 196os for
example, it was reasonably estimated that to provide adequately for the legal
needs of the poor, 49,000 legal services lawyers and public defenders would
have to be employed. At the time there were perhaps 4,000 attorneys, both
volunteers and quasi-governmental employees, who were serving in those
capacities. 27
To comprehend the relationship between resources and practice, one
must understand the meaning of demand in public services. Demand is not
only part of a transaction between citizens and government but is also a
transactional concept. It not only requires a demander but also a more or less
encouraging supplier. It is meaningless to specify a level of demand unless
34
The Problem of Resources
this is accompanied by a positive (if implicit) degree of receptivity. 28 In more
concrete terms, demand for services can be estimated but it cannot finally be
known in the abstract. It is a function not only of expressions of client prefer-
ences but also of government efforts to offer services and to record or ac-
knowledge client responses. Demand as an expression of desire for services
is a meaningful concept only if it is accompanied by explication of the extent
to which demand was sought out.
Overt expressions of demand for services tend to be more responsive to
changes in the perceived availability of services than in changes in the un-
derlying conditions that are commonly supposed to affect demand. In other
words, perceived availability of service "pulls" demand, not the other way
around. To illustrate: when cities introduce "gu" programs providing for
central reception of public safety telephone calls, an easily memorized tele-
phone number and publicity to instruct the public in its use, they discover
an untapped reserve of requests for public safety assistance. Indeed, New
York City at one point undertook a public campaign to urge people to use the
911 system less because the increased demand outstripped the city's ability
to respond. When Boston introduced its Little City Halls program, provid-
ing a municipal presence in the neighborhoods, requests for city services
through complaints increased by a third. 29
Service programs generally can increase the recorded demand on them-
selves by changing the nature of outreach, the location of offices, or public
information about the program. 30 Putative demand for services cannot even
take shape until the structures for receiving demands are in place. For ex-
ample, in Boston 4,000 students were assigned to classes for the retarded in
a recent year, whereas only 70 were classified as emotionally disturbed.
These figures might suggest that there was enormous demand for special ser-
vices for the retarded in comparison to the needs of emotionally disturbed
pupils. This situation would surprise most specialists in learning disabilities
because, to the extent that these labels have some objective referents, these
conditions tend to occur in about equal proportions in the population (about
2 percent). It would seem a reasonable inference that the disproportionate
demand for services for the retarded was not intrinsic among Boston youth
but rather was entirely a function of the availability of special classes for one
group and not the other and of the preferences of Boston school personnel
for assigning pupils to this category of disability. 31
The proposition that demand will increase to meet the supply applies
qualitatively as well as quantitatively. If there were a fixed clientele (we have
just argued there is not) clients would still demand more and improved ser-
vices, as the population has done historically. This might seem like the ideal
35
CONDITIONS OF WORK
37
CONDITION S OF WORK
such ways are so high as to be politically out of the question except in the
case of tantalizing, carefully managed pilot or demonstration programs that
provide hints of what social welfare delivery might be like if service quality
were given primacy. Typically, if such programs succeed, their cost-be.1efit
ratios are regarded as unacceptable and pressures develop to provide service
in less costly ways.
This analysis of the demand-supply dilemma should not be taken as
counsel of despair. Public policy always requires consideration of the trade-
offs involved in providing additional resources for added benefits and incur-
ring additional costs. With added resources more people can be served, just
as more people can get to New York City from Long Island, although under
stressful conditions, by using an expanded Long Island Expressway. But ap-
preciation of the demand-supply dilemma in street-level bureaucracies does
suggest that the problem of the quality of service delivery is not likely to
yield easily to any imaginable resource increments. Other things being
equal, increased capacity results in reproducing the level of service quality
at a higher volume for any imaginable increase in resource availability. This
proposition is critical because it explains why the steady increase in re-
sources available to street-level bureaucracies in recent years has not re-
sulted in improvements in the perceived quality of client treatment. (Other
reasons include the fact that salary raises, which consume increases in
agency resource allocations, do not increase resources available to clients, al-
though they may help to maintain staff quality.) Further, it contradicts many
often self-serving perspectives on reform, which hold that additional person-
nel are the most important ingredients in responding adequately to citizen
complaints.
Thus street-level bureaucracies are often trapped in a cycle of mediocrity.
The better the program and the more responsive it is to the needs of the citi-
zens, the greater will be the demand for the service. This larger demand
forces the agency to limit service artificially or impose costs on clients in the
absence of a pricing mechanism that would otherwise ration service. The im-
po'ied cost of inferior quality or difficulty in receiving services continues
until, in the extreme, the agency is returned to the previous equilibrium of
indifference to client needs. The more successful the organization, the more
likely it is to encounter this dilemma.
It is generally characteristic of free government programs that demand
will increase to the point that goods or services can be provided. And in
response to this demand agencies providing the goods or services will im-
pose monetary or nonmonetary costs in order to limit demand effectively. 39
Street-level bureaucracies also develop rationing mechanisms to impose
The Problem of Resources
costs for services but in doing so they face certain constraints. If indeed they
provide vital services-income to the indigent, public safety, education,
health care-then there are severe constraints on the sorts of rationing
mechanisms they can impose.
Street-level bureaucracies must not appear to be rationing services or
depriving social groups of their rights or entitlements. They must give evi-
dence of strenuous efforts to avoid reducing services. They may be asked to
"trim the fat," but never to reduce the quality of services or the quantity of
vital services. It is one of the best-kept secrets in government how agencies
can forever find fat to trim and nonessential services to eliminate, while
never affecting "vital programs" and "necessary services."
Agencies forced to reduce expenditures significantly, or otherwise limit
service provision, typically publicize the dire consequences of failure to spe-
cific groups or sectors to provide budgetary support for what they are doing.
(For example, the public university that announces it will be unable to admit
a freshman class.) But when it comes to making cuts they will typically at-
tempt to reduce services "across the board" rather than severely injure spe-
cific sectors of the population.
The demand-supply dynamics of street-level bureaucracies provide addi-
tional understanding of why they seem to be chronically understaffed.
The insatiable demand for services in basic social programs results in agen-
cies characteristically responding, and being forced to respond, by adding
services or clients, or extending existing resources under pressures of
budget stringency. The ethics of public service requires that they provide
more of what they do or give rather than utilize additional resources to
improve the resource-demand balance.
Sometimes agencies enjoy freedom from close scrutiny on spending. Then
they can add qualitative improvements or reduce case-load pressures, as edu-
cators have been able to do from time to time. Or, like the police, their ser-
vices may be in high demand, and they can ride public concerns to new
levels of spending for men and machines.
But for the most part street-level bureaucracies, as major components of
local public spending, cannot long escape close budgetary scrutiny and thus
become the targets of the taxpayers' revolt. In an inflationary era the budget-
ary convention of asking agencies to maintain effort on the basis oflast year's
allocation, while costs for goods, services, and salaries are rising, insures that
street-level bureaucracies will experience their resource position as chroni-
cally inadequate.
39
CHAPTER 4
Goals
43
CONDITIONS OF WORK
44
Goals and Performance Measures
treatment versus routinization and mass processing, and response to the
needs of individual clients versus efficient agency performances.
These dilemmas are related to the public nature of the programs. Just as
agencies distributing free goods must develop mechanisms to ration their
allocation, political systems must place limits on the demands that organiza-
tions can make for additional resources. If, indeed, demand will increase to
equal the supply then the inherent impulses organizations display toward
growth 11 will lead to increasing organizational scope even though, as sug-
gested above, the quality of service cannot be expected to improve with
growth.
Suggestive evidence for this proposition is encountered in the alarm that
public officials express when they discover that benefit programs are open-
ended. The food stamp program, for example, caused a great deal of official
consternation because it entitled many more people to benefits than was
originally anticipated, resulting in unexpected costs difficult to control ex-
cept through categorical cutbacks. 12 Constant harping on the error rate in
public welfare or allegations of doctor abuses of medicare claims serve to
remind the public and the agencies in charge of these programs that uncon-
trolled growth in government spending is not officially acceptable. Thus, ex-
cept in rare instances, such as a new service program in search of a clien-
tele, 13 street-level bureaucracies are under continuous pressure to realize
the public objectives of efficiency and cost effectiveness. Pressures will be
more or less explicitly articulated depending upon the political climate and a
variety of other factors.
Like other organizations, street-level bureaucracies are always con-
strained by resource limitations. However, it is important to understand that
this constraint is experienced as a source of tension within street-level bu-
reaucracies at the operational level. Street-level bureaucrats must find a way
to resolve the incompatible orientations toward client-centered practice on
the one hand and expedient and efficient practice on the other.
45
CONDITIONS OF WORK
and ambiguity (note that goals are one dimension of role construct). There
are at least three ways in which the complicated structure of street level bu-
reaucrats' role expectations contributes to goal ambiguity and conflict.
First, to the extent that public expectations affect street-level bureaucrats
there is often considerable disagreement about what street-level bureau-
cracies should primarily do. Street-level bureaucrats within limits may de-
fine the ways in which they will pursue their objectives. But often commu-
nity opinion is diffusely apprehended, creating role conflict. In different
cities and within the same city some communities may expect police to en-
force the law vigorously and impartially and to have a strong legalistic orien-
tation, while others are content to have the police primarily focus on main-
taining order. Some parents want schools oriented toward achievement in
basic skills, while others want schools oriented toward developing commu-
nity values and vocational training. The punitiveness of welfare administra-
tion in some cities and the liberal orientation of other welfare offices demon-
strate clearly how the same statutes can spawn agencies with vastly different
bureaucratic cultures. 15
These differing perspectives on the purposes of service contribute to goal
uncertainty and lead to the following hypotheses. To the extent that com-
munities are indifferent to the nature of bureaucratic policy or fail to express
their views in politically salient ways, street-level bureaucracies will perform
with internally generated objectives. Conversely, the stronger community
sentiment is concerning proper bureaucratic behavior, the more street-level
bureaucracies will respond to community orientations. The more heteroge-
neous community sentiments are, however, the more street-level bureau-
cracies will experience goal conflict.
Major urban conflicts have resulted in recent times over such diverse
perspectives on public services. Should schools promote appreciation of eth-
nic and racial heritage or should they be devoted to the traditional role of
homogenizing and Americanizing urban newcomers? Should public housing
managers attempt to integrate their projects racially or should they try to
maximize the security and comfort of long-term project residents? Is dis-
cipline and punishment more effective for inducing studious attitudes among
pupils or is a liberal and flexible approach more appropriate? These are ques-
tions that would be answered quite differently by different subpopulations of
the country's largest cities. 16
Racial conflict brings issues of divided community sentiment into sharp
focus. Black community residents differ from whites in their law enforce-
ment priorities, the desirability of police intervention in certain areas, ap-
propriate role models for their children in schools, and so on. Moreover, res-
Goals and Performance Measures
idents of black communities themselves differ in priorities; for example,
whether they would prefer high-intensity police protection in order to
reduce crime or whether they regard saturation patrolling as community
harrassment .
A second dimension of role conflict or ambiguity stems from the significant
role of peer groups in establishing role expectations. For street-level bureau-
crats, peers are fellow workers (although generally peers can be otherwise,
e.g., social peers, family peers, etc.). Only work peers fully appreciate the
pressures of work and the extent to which street-level bureaucrats experi-
ence the need to have goal orientations that are consistent with resolving
work pressures. The greater the strain between various goal expectations,
and the smaller the zone of indifference in which street-level bureaucrats
operate, the more peer support is critical for sustaining workers' morale. 17
The subject of peer support is discussed extensively in analyses of the
police, the street-level bureaucracy that is perhaps the most controversial
and the most subject to conflicting goal expectations. Police must perform
their duties somewhere within the demand for strict law enforcement , the
necessity for discretion in enforcemen t actions, and various community in-
terpretations of proper police practice. They must accommodate the con-
straints of constitutional protection and demands for efficiency in mainte-
nance of order and crime control. They must enforce laws they did not make
in communitie s where demands for law enforcemen t vary with the laws and
with the various strata of the population. 18 Police may perceive the public in
these communitie s as hostile yet dependent. Police role behavior may con-
flict significantly with individual value preferences and the behavior and
outlook of others with whom police must work, particularly judges. Police
are expected to be scrupulously objective, impartial, and upstanding, pro-
tective of all segments of society even while society does not protect all seg-
ments of itself. 1e
A third dimension of the construction of street-level bureaucrats' role ex-
pectations concerns the role of clients. Clients are not a primary reference
group of street-level bureaucrats. They do not count among the groups that
primarily define street-level bureaucrats' roles. 20
This is not to say that children are unimportant to teachers or that litigants
and defendants are unimportant to judges. But these people do not primarily
or even secondarily determine bureaucratic role expectations. Work-relate d
peer groups, work-related or professionally related standards, and public ex-
pectations generally are much more significant in determining role behavior.
Recognizing their weak influence in defining workers' roles, some client or-
ganizations have demanded inclusion in the constellation of bureaucratic ref-
47
CONDITIONS OF WORK
Performance hleasures
49
CONDITIONS OF WORK
high risk. Being poor, black, and a student at an inferior high school was
regarded as high risk enough. Programs often admitted students with rea-
sonable grades, neglecting students who might have been even higher risks.
If those programs succeeded did they succeed because they selected stu-
dents prudently, or because they ran effective programs? Moreover, evalua-
tion would have to consider changes in the social or economic environment.
Changes in national employment trends or civil rights orientations might af-
fect program success in ways not attributable to the program.
To some degree public deference to street-level bureaucrats' autonomy in
decision making is also characteristic. This deference is a defining aspect of
professionalism and has some applicability in all the areas in which street-
level bureaucrats work. While public deference is related to the needs for
discretionary judgment, goal ambiguity (leaving goals to be defined by pro-
fessionals), and decision complexity, it has a meaning of its own, particularly
as public bureaucracies attempt to develop performance measures as a pre-
condition to increasing bureaucratic control.
Street-level bureaucrats tend to perform in jobs that are freer from super-
visory scrutiny than most organizational jobs, and work norms prevailing in
these jobs minimize such scrutiny. This may be because of supervisors'
willingness to respect professional claims (for example, teachers in most
schools are rarely visited in classrooms by principals, and then only with
enough notice so that performances can be staged). Or it may be because
workers expected to exercise discretion require some freedom from super-
visory control (for example, policemen whose supervisors leave them alone
as long as they seem to be performing adequately). 25 Whatever the sources
of this freedom, it contributes in itself to the problems of measuring perfor-
mance, particularly since peer evaluation is one of the ways to achieve ac-
countability in work quality.
Notwithstanding these difficulties, however, bureaucracies do establish
standards and measure workers' performances against these standards. For
example, policemen typically are asked to make a certain number of arrests
per month. Social workers are asked to maintain a certain monthly intake
and case-closing rate. But these measures are only problematically related to
public safety, or to clients' ability to cope with problems that are in part the
objectives of these interactions. And they have nothing to do with the appro-
priateness of workers' actions, or the fairness with which they were made,
the net results of which determine the rates on which workers are judged.
Not only are such standards problematically related to goals, but it is not
even apparent whether measured increases or decreases signal better or
worse performance. The best illustration of this consideration is the problem
so
Goals and Performance Measures
of inference from crime statistics. Do increases in arrest rates signal im-
proved police performance? Or do they signal deteriorating police perfor-
mance, indicating an increase in criminal activity and thus in the number of
criminals available to catch? Actually, changes in arrest rates may indicate
neither, reflecting rather changes in the focus of police patrol. Do decreases
in the welfare rolls signal improved bureaucratic performance? They do only
if improvement is defined as having fewer people on welfare, the economy
that creates the need for welfare remains stable, and welfare agencies do
not change their practices affecting clients' willingness to apply. These are
but specific illustrations of the observation that agency-generated statistics
are likely to tell us little about the phenomena they purport to reflect, but
a great deal about the agency behavior that produced the statistics. 26
Despite the difficulties of performance measurement, street-level bureau-
cracies do seize on some aspects of performance to measure. They tend to
seek reports on what can be measured as a means of exercising control. In
tum, the behavior of workers comes to reflect the incentives and sanctions
implicit in those measurements.
The relationship between performance measures and behavior was per-
haps first highlighted by sociologist Peter Blau when he observed that when
the employment agency he WllS studying began to be evaluated in terms of
its placement rate, employment counselors shifted the focus of their work to
the more easily employed at the expense of those more difficult to place. 27
This illustrates the general rule that behavior in organizations tends to drift
toward compatibility with the ways the organization is evaluated.
Street-level bureaucracies also measure the training and experience of
employees as a way of assessing quality. These surrogates represent qualities
that are hypothetically associated with good performance. Thus a teaching
faculty is assessed in terms of its educational background although the gradu-
ate training in question is only problematically related to effective teaching.
Street-level bureaucracies often depend on the experience or training of
their workers as signs of quality service, although it is not clear that more
training or experience is associated with doing a better job. Low-income
parents often express concern that experienced teachers exercise seniority
rights and migrate to schools in -affiuent neighborhoods. But experience is
problematically related to effective teaching. The less-experienced teacher
may be more interested, energetic, ambitious, more recently educated and
familiar with new techniques, and more sympathetic toward a low-income
clientele. (Of course, lacking good performance measures we rarely have an
opportunity to test the relationship between the surrogate measures and
their implications for performance.)
51
CONDITIONS OF WORK
52
Goals and Performance Measures
An important problem of public bureaucracies generally and street-level
bureaucracies particularly is that clients do not receive the kind of informa-
tion that would permit them to compare or assess their treatment. Nor can
they compare the treatment they receive this year with the treatment ex-
tended to clients in other years, or compare the performance of their agency
with similar agencies elsewhere. Citizens in general and poor people in par-
ticular will resign themselves to inferior levels of service if they have nothing
with which to compare their experiences and have no basis for thinking that
they deserve any better. Their frame of reference, if any, is experiential. But
the isolation of most clients from each other makes it difficult to interpret ex-
periences effectively and makes clients highly subject to street-level bureau-
crats' definition of their situation.
Recent experiences with educational voucher experiments illuminate the
importance of the absence of performance measures in street-level bureau-
cracies. First, teachers were very resistant to comparison to other teachers
and to the staffs of other schools. Second, parents rarely exercised their op-
tion to shop around for the best school, tending instead to support the con-
ventional schools. From the standpoint of the present analysis one could
hardly expect parents to exercise options when they had no basis for under-
standing which schools performed better. Whatever problems they may
have had with their schools parents may have attributed to difficulties within
themselves or their children. In the absence of solid information on alterna-
tives the security of the known quantity was in any case generally chosen
over the unknown quantity. Significantly, in the experiments in which
school systems were able to differentiate the programs of each school,
parents did exercise considerable options. 35
In summary, the inability to measure street-level bureaucrats' perfor-
mance has widespread implications for controlling the agencies. Supervisors
and agency directors can discipline workers, but not to the point of closely
guiding workers' activities toward agency preferences unless they can moni-
tor performance and determine who is or is not measuring up. Nonetheless,
surrogate performance measures are developed to provide the agencies and
the public with control tools, even if the tools are not quite appropriate and
may even be counterproductive to the purposes. These measures do affect
workers' behavior, although not necessarily in the direction favored by the
agency or the public. And street-level bureaucrats, in recognition of the im-
portance performance measures have to limiting their autonomy, actively
resist their development and application.
53
CHAPTER 5
Nonvoluntary Clients
54
Relations with Clients
poor clients provide different treatment in comparison to those serving more
affluent people entirely on the basis that poor people are much more depen-
dent on these agencies. 1
What is the meaning of saying that the clients of street-level bureaucracies
are nonvoluntary? What difference does it make and why do we elevate this
consideration to the status of a primary working condition?
If street-level bureaucracies have nonvoluntary clients then they cannot
be disciplined by those clients. Street-level bureaucracies usually have noth-
ing to lose by failing to satisfy clients. They will try to manage a large volume
of complaints and undoubtedly seek to minimize the extent to which they
are perceived as difficult to deal with or unresponsive. But managing com-
plaints successfully is a far cry from changing policy in response to consumer
dissatisfaction. Yet, as indicated in the previous chapter, receiving com-
plaints and correcting policy in response to them is one of the few ways orga-
nizations can learn from clients. 2
Sometimes street-level bureaucracies are even rewarded for reducing
their clientele. Public welfare agencies in the United States come under crit-
icism for their large number of clients, although a public agency that
reached less than half the eligible population in other contexts might be
regarded as a colossal failure. At other times street-level bureaucracies are
indifferent to the loss of clients or client dissatisfaction. Partly this is due to a
proposition developed earlier. If demand for services is practically inex-
haustible relative to supply, then the fact that some clients are disaffected by
the quality or level of service means only that their places are taken by
others who need the service and are willing to accept the costs of seeking it.
Even in situations in which one would assume that street-level bureau-
cracies would suffer from client losses, the bureaucracies often seem rela-
tively indifferent. Consider the case of public schools in central cities that
have lost white, middle-class students and failed to retain lower-class stu-
dents. Although school revenues are based on average, per-pupil attendance
these population losses usually have little effect on school practices.
Perhaps there are ways to explain this lack of responsiveness. The tenure
of the more experienced teachers, for example, may mean that they do not
consider their jobs threatened by declining school enrollment. Or perhaps
school officials think that they are not capable of combatting the attractions of
suburban and parochial schools or of influencing achievement levels among
drop-outs. Whatever the full explanation of schools' unresponsiveness to
declining enrollments, 3 they place surprisingly little emphasis on programs
to retain pupils and on public relations designed to promote their strengths
55
CONDITIONS OF WORK
and enhance their long-run budgetary status. Instead schools fight for their
budgets within the arena of local-interest group politics without great em-
phasis on consumer satisfaction.
That clients are nonvoluntary has significant implications not only for the
direction of public services as a whole but also for the quality of interactions
between street-level bureaucrats and clients. Primarily, this is because non-
voluntary clients cannot avoid or withdraw from encounters with workers.
Where both parties are free to continue the interaction or leave it, partici-
pants will set limits to the costs they will accept before ending the rela-
tionship. If the encounter is instrumental, that is, if each participant wants
something from the other, they will continue to pursue their objectives
within the relationship so long as they value the objectives more than the
cost of seeking them. This permits a wide range of implicit bargaining tac-
tics, particularly if both parties have a stake in maintaining the relationship. 4
However, if one of the parties does not enter the relationship voluntarily
or must sustain the relationship because a highly desired good for which
there is no alternative is controlled by the other person in the encounter, the
nature of the interaction changes. The costs that the non voluntary person in
the interaction will sustain become much higher. Indeed, the less voluntary
the interaction, the less useful it is even to understand the interaction in
terms of limits to the costs people will accept, because clients cannot easily
withdraw.
Street-level bureaucrats can impose costs of personal abuse, neglectful
treatment, or inconvenience without necessarily paying the normal penalty
of having the other party retaliate. When medical personnel refer to patients
as "garbage," "scum," "liars," "deadbeats," and so forth, there is a tempta-
tion to say that this is a reaction to the moral superiority they feel over lower-
class people. 11 However, neglect and abuse of patients is a function of the
nonvoluntary nature of the association of clients with patients, and not
strictly of bureaucracy or class discrepancies. This is suggested by the obser-
vation that doctors in private practice can also neglect patients if there is a
restricted supply of professionals, but they become much more solicitous
when patients have medical alternatives on which to draw. 6
In the elliptical euphemism used by several street-level bureaucracies,
they need not worry that clients will"elope." If clients refuse to continue in-
teracting with street-level bureaucracies, the fault may always be attributed
to the client. "Escapees," "dropouts," "incorrigibles," and "socially disorga-
nized" are labels that imply that the exit of the client is attributable to a
defect of the client.
Relations with Clients
The nonvoluntary nature of clients helps explain why they are not among
street-level bureaucrats' primary reference groups (see chapter 4). But this
does not mean that clients are helpless in the relationship. Street-level bu-
reaucrats in a sense are also dependent upon clients. Clients have a stock of
resources and thus can impose a variety of low-level costs. This is because
street-level bureaucrats must obtain client compliance with their decisions,
particularly when they are evaluated in terms of clients' behavior or
performance.
Order in a prison is a function of adjustments made by guards in exchange
for prisoners' general compliance with regulation. 7 So it is with most social
organization. To do their work smoothly police officers must obtain the con-
sent of suspects they apprehend. Teachers must secure pupils' cooperation
before they can begin to teach. Social workers must obtain the compliance of
welfare recipients in case determinations or confront time-consuming ap-
peals. The child who refrains from asking a question after a certain point out
of fear of making the teacher angry, the traffic law violator who fails to con-
test what he regards as an officer's unfair decision, the law client who, al-
though vaguely troubled, is too confused to raise questions about her case,
are all actively cooperating in the interactions.
For the most part, except in the more coercive bureaucracies, clients give
their consent because (sometimes in combination) they accept the legitimacy
of the street-level bureaucrats' position and decision, anticipate that dissent
would not be productive, or consider themselves favored by the decision or
action taken. Most encounters with bureaucracy appear to be characterized
by the consent of clients, but the structure of choices available to clients
limits the range of alternative behaviors that they consider realistically avail-
able. In short, clients' consent is continuously being managed by public
agencies.
Street-level bureaucrats are not required to command. Clients control
themselves in response to the superior power of the workers. This is not to
suggest that clients are docile because swift retaliation would result from
noncompliant behavior. Rather, compliance in most street-level bureau-
cracies may be said to result from the superior position of the workers, their
control over desired benefits, and their potential capacity to deny benefits or
make their pursuit more costly.
Compliance also results from the milieu, which comprehensively cues
clients concerning behavioral expectations. Readers lower their voices in
57
CONDITIONS OF WORK
libraries, defendants talk respectfully to judges, patients wait quietly for doc-
tors, children obey teachers, not directly because of the imminence of retali-
ation but because they have a diffuse appreciation of "proper" modes of be-
havior and a diffuse awareness that deviance from these norms may be
punished. 8
Nonetheless, street-level bureaucrats sometimes do display behavior that
strongly suggests this inference is warranted. Street-level bureaucrats in-
deed reprimand or otherwise sanction deviance from acceptable standards of
client behavior. They dominate their interactions with clients. They cue and
otherwise teach clients to behave "properly." They structure work patterns
to maximize control over clients independent of any policy objectives.
The need to control clients is a requirement evident in many work areas,
not just street-level bureaucracies. Relationships are always reciprocal to
some degree. If one party seeks to control the other, the second party may
increase the costs of the first party gaining or exercising control, even if the
first is unquestionably more powerful. This observation, which has universal
applicability from guerrilla warfare to concentration camps, takes particular
shape in street-level bureaucracies in several respects.
First, street-level bureaucrats characteristically are pressed with heavy
case loads and demands for quick decisions, so that clients can impose salient
costs merely by taking workers' time. Since time may be fairly cheap for
clients, or their needs high relative to the value they place on their time,
clients potentially have a store of resources with which to affect their rela-
tionships with street-level bureaucrats.
Second, street-level bureaucrats are characteristically constrained in the
resources they can employ in obtaining client compliance. These constraints
consist of professional and bureaucratic standards of fairness and due process
that to some degree place limits on what can or cannot be done to or with
clients (notwithstanding the most outrageous tales of exceptions to the con-
trary). They are also constrained by social norms of proper behavior toward
other people and by recognition that power should be accompanied by re-
sponsibility, particularly when clients are identifiably (indeed defined as)
socially or economically needy. This point is emphasized not because street-
level bureaucrats are absolutely constrained from abusing their positions,
but because what needs to be explained is the mobilization of control in com-
bination with constraints against excessive manifestations of power. Modern
bureaucracies which are too heavy-handed lose their legitimacy if their of-
fenses are publicized. Moreover, they are ultimately inefficient if they
require significant force to assure adequate client control.
Third, there is an extent to which clients' satisfaction or performance is
ss
Relations with Clients
important to street-level bureaucrats. Successful intervention, expressions
of gratitude, and changes in behavior in the desired direction are valued by
street-level workers whether or not these developments are reasonably at-
tributable to their work.
Clients sometimes manipulate the gratification received by street-level
bureaucrats in order to affect future interactions. Client strategies include
passivity and acquiescence, expressions of empathy with workers' problems,
and humble acceptance of their own responsibility for their situation. The
disadvantaged position of clients forces them to conspire in their own man-
agement in order to avoid offending the workers or providing negative evi-
dence about their character. In some circumstances clients can effectively
express anger or demand their rights, but these strategies appear useful only
in certain circumstances and usually not for long. 9
While a client has some resources with which to affect a relationship with
street-level bureaucrats, the relationship is by no means a balanced one. It is
a relationship of"unidirectional" power in which "the capacity to make and
carry out decisions is the exclusive, or near exclusive, property of one of the
... groups." 10 The relationship is primarily determined by the priorities
and preferences of street-level bureaucrats, but the character and terms of
the relationship are substantially affected by the limits of the job.
59
CONDITIONS OF WORK
though they come from different class backgrounds (even here we need to be
cautious, since fighting may mean something quite different to people of dif-
ferent classes). Likewise, the boxers may have a similar view of reality be-
cause they share the common goal of putting on a show, or because, while in
conflict, they are relatively evenly matched.
There is little agreement, however, on the picture of reality where clients
and street-level bureaucrats are concerned. This is at least because the two
are intrinsically in conflict over objectives and the relationship is drastically
unequal. What street-level bureaucrats think they do may have little con-
nection with what clients think is going on. Clients tend to experience their
needs as individual problems and their demands as individual expressions of
expectations and grievances. They often expect treatment appropriate to
them as individuals, and are in large measure encouraged in this expectation
by public institutions and society in general. On the other hand, street-level
bureaucrats experience client problems as calls for categories of action. Indi-
vidual client demands are perceived as components of aggregates. Expecta-
tions of proper treatment are framed in terms of satisfactory solutions for the
optimal processing of the totality of the work rather than in terms of the best
solution for individual cases. Clients seek services and benefits; street-level
bureaucrats seek control over the process of providing them.
There are four basic dimensions to the control exercised by street-level
bureaucrats over clients. Each significantly affects some dimension of client
"construction." Briefly, street-level bureaucrats exercise control in: (1) dis-
tributing the benefits and sanctions that are supposed to be provided by the
agencies; (z) structuring the context of clients' interactions with them and
their agencies; (3) teaching clients how to behave as clients; and (4) allocating
psychological rewards and sanctions associated with clients entering into
relationships with them.
6o
Relations with Clients
vices under agency policies. It may be because classifying the behavior or
background of the client is a matter of discretion, as in assessing whether the
behavior of a pupil or prisoner constitutes insolence. Or it may be because
the categories into which clients fit are actually problematic and not fixed, as
in judicial determination that a juvenile defendant is" a basically good kid in
trouble" rather than "a rotten apple. " 12 To the extent that the assigm;nent of
benefits or sanctions is negotiated between street-level bureaucrats and
clients through interpersonal strategies and implicit maneuvering, the allo-
cation of benefits and sanctions are clearly part of a process of constructing
the client profile.
1g64-1965, tenants were able to gain priority with local housing agencies
because of their association with rent strike organizations. 17 "Nothing more
can be done" often really means: "Priorities will not be changed in your case,
although they could be." Since priorities could be and often are changed
frequently, bureaucracies have a stake in concealing the mutability of policy.
(It may be that bureaucratic managers often do not themselves believe or un-
derstand that priorities could be changed.)
At the agency level, bureaucracies also attempt to convey proper levels of
expectations. Long lines not only discourage prospective clients but also
convey that many people have to be processed; hence individual clients
should appreciate that workers have little time to spend with them or on
their problems. Conducting welfare intake interviews in a single intake of-
fice so that workers and clients can overhear all the interviews being con-
ducted conveys to clients that they may not expect privacy. Lower-court ar-
raignment sessions clearly convey to waiting defendants that the judge is
very busy, that the court has developed procedures to speed the process,
that the court appreciates cooperation with the process, and that most peo-
ple accept their inability to understand what is going on.
Street-level bureaucrats often attempt to involve clients in the difficulties
of their jobs in order to gain understanding or sympathy for their position.
Assertions that ''I'm just doing my job," or ''I'm following orders" help bring
the client to an agency point of view. The client is implicitly asked to aban-
don his or her own interest in the interaction in a friendly, not overtly
conflictual tone. But there is little choice involved, since the structure of the
institution requires the client to comply or else risk alienating the more pow-
erful street-level bureaucrat. No doubt many workers genuinely assert such
claims, and they are often genuinely appreciated by clients. Still, in poten-
tially conflictual stuations these claims function to gain client compliance in a
persuasive rather than coercive way.
Fourth, under some circumstances street-level bureaucrats convey infor-
mation about how to work the system. In so doing, they alert clients to the
alternatives available to them under the current structure. Most clients
would like to know more about how to negotiate the system, but this infor-
mation is rarely provided to all clients. Rather, street-level bureaucrats exer-
cise discretion by providing this information on a selective basis. This be-
comes one of the few ways they are able to favor clients without directly
abridging bureaucratic norms of fairness. They make no decisions in favor of
one client over another. They simply inform clients selectively how to utilize
the system to best advantage. Thus they respect fairness in decision making;
it is only information that is selectively distributed. Clients who are so fa-
Relations with Clients
vored, however, receive a tremendous benefit when norms of universalistic
decision making operate. Knowing how to position oneself so as to increase
the probability of a favorable decision is a substantial advantage. 18
Sometimes teaching clients how to work the system consists of favoring
some clients by providing them with special information. Public housing
officials in Boston did this in coaching elderly applicants on how to apply
for emergency housing status. 19 Boston housing inspectors did this by re-
minding landlords that they could petition for certain hearings. 20 Sometimes
teaching the system takes the form of discriminating against some clients by
denying them information given to others. Welfare officials in North Caro-
lina apparently did this in informing most but not all clients of their right to
apply for assistance. 21
At other times street-level bureaucrats teach clients how to get best re-
sults from other bureaucracies. This practice is common among those work-
ers who prepare clients for other agencies. Thus a social worker will instruct
a client how to obtain favorable treatment in a referral. A lawyer will instruct
a client how best to impress a judge. 22
Coaching selected clients at times seems innocuous, but it almost always
has redistributive effects. The defendant instructed to appear penitent
makes others who fail to receive coaching appear defiant in contrast. The
public housing officials who coached elderly applicants on how to fit into
emergency categories helped these applicants to receive better units more
quickly, to the disadvantage of other, equally deserving applicants. Prospec-
tive welfare recipients were harmed by the failure of welfare officials to
explain their right to apply for services. 23
To the extent that biased coaching remains secret, one would hypothesize
that public legitimacy and regard for the agency would increase. No one
would perceive himself or herself to be worse off, while some would feel
favored. But to the extent that these biases become known or are exposed,
they would tend to undermine the agency's legitimacy.
regard for each other. 24 When one person in an interaction has status and
power relative to the other, the signals emanating from that person are par-
ticularly potent. Since a person's self-concept is substantially a function of
the response of others who are important to the person, interactions with
street-level bureaucrats have psychological as well as material implications.
The psychological implications of interactions with street-level bureau-
crats may be fleeting where the interaction is not sustained. Police tend to
treat people they apprehend scornfully or respectfully depending upon their
apparent moral worthiness and the respect they display for the police. To be
treated with respect, or with utter disrespect, by these symbols of authority
have implications for citizens' views of themselves. People stopped by the
police in a sense discover whether they are or are not the kind of person to
whom respect is normally granted.
Frequently, citizens do not understand why police have stopped them or
singled them out. They attribute an officer's brusque and imperious manner
to one of their own personal or physical traits. If they are members of minor-
ity groups they may conclude that their racial or ethnic identity triggered
police intervention, confirming once again the hazards of minority status and
their belief that police officers are racist.
Anger at bureaucrats in general may be attributed to the dehumanizing
aspects of having to seek service through bureaucracy. However, for most
people apprehension by the police, visits to emergency rooms, or court ap-
pearances are single events which have limited implications for personality
development. But any of these events may lead to further bureaucratic en-
tanglement, may reinforce patterns of interaction previously encountered,
or may signal other bureaucracies that the same (damaging or rewarding) ori-
entations toward the client should be forthcoming in the future.
The greater the involvement of the client with the agencies and their em-
ployees, the more sustained and critical the psychological implications of the
interactions. For this reason the dependence of poor people on government
services creates the context in which interactions with street-level bureau-
crats may have substantial psychological implications. At the very least poor
people who bounce from one agency to another have reinforced feelings of
dependency, powerlessness, and, deriving from these, anger. After sus-
tained exposure to the welfare system, for example, recipients have been
found to see themselves as "undeserving" and "lucky to get anything at
all. "25
Institutions that fully dominate peoples' lives have extensive influence
over personality development. As Erving Coffman has demonstrated, men-
tal hospitals teach patients how to be patients by rewarding behavior that
66
Relations with Clients
conforms to staff expectations of how mentally ill people behave. Thus they
not only teach the client role but touch the person playing the role as well,
since for mental patients the role is also their salient personal identity. 26
The closer institutions get to total involvement with clients, the more
their self-images may be affected in a sustained way. Therapists and counse-
lors utilize this fact to reinforce positively the self-images of their clients.
Through residency in supportive halfway houses, peer support groups, and
other environmental influences, some therapists seek not only to help clients
develop psychologically, but also to insulate them from contradictory imag-
ery during the period of involvement. 27
For most people, the street-level bureaucrats whose psychological reac-
tions are most powerful are teachers. Teachers powerfully convey images to
children concerning their expectations of achievement. These images affect
the child's self-image, self-expectations, and actual achievement. Through
formal, and more insidiously, through informal tracking, teachers indicate to
students who is expected to achieve and who is not.
In terms of the outcomes of institutional involvement with citizens, the
policy problem is not only that students labeled as likely or unlikely to
achieve do perform at expected levels, thereby fulfilling the original achieve-
ment prophecy. 28 It is also that determinations concerning the likelihood of
success are not even made on the basis of achievement potential, measured
however crudely, but are made on the basis of inferred social class, or the
biased "harder" data accumulated or reported by other street-level
bureaucrats.
In a careful study of a group of primary school pupils, Ray C. Rist ob-
served that kindergarten pupils were placed in putative ability groupings
solely on the basis of their dress, demeanor, verbal skills, and social back-
ground, before any sustained interaction between the teacher and the chil-
dren and without testing. 29 The fast-track children received most of the
teacher's instructional attention and most of the classroom rewards, although
performing honorific chores did not require cognitive skills. The group
placed in the lower classroom tracks learned from the teacher that they were
not expected to achieve and that they were the kinds of children who could
be ignored and ridiculed by authoritative adults (the teacher) and their peers
with higher alleged learning ability (the children in the fast track). Mean-
while, the fast-track students were instructed in :everal ways about their
proper self-image. They learned not only that they were brighter than the
other children, but also that it was legitimate to scorn the others.
Rist noted that in subsequent grades the relatively poor performances of
the slow-track children (perceived as such by the teacher, who had insured
CONDITIONS OF WORK
that it was so) were reified into fact and formed the basis for other teachers'
tracking placements. He also found indications that the lower-track children
were learning the material, despite the fact that the teacher tended to ignore
them. The teacher did not know this because she did not encourage the
lower-track (and lower-class) children to display their knowledge, and the
cumulative effects of the classroom tracking system discouraged them from
displaying their knowledge.
Analogous lessons are communicated in interactions in other street-level
bureaucracies. Their potency depends on the extent to which clients have
similar dependency and sustained interactions. The reification of street-level
bureaucrats' prior judgments in subsequent placement decisions is also com-
mon. Judges' decisions concerning severity in sentencing juveniles, for ex-
ample, depends substantially upon the defendant's record rather than on the
severity of the offense. 30 Disposition of students with behavioral disabilities
also depends on the previous incidents recorded. 31 In both these instances it
is easy to see that young people are subject to the hazard that in the past they
may have encountered adults who regarded their behavior as requiring of-
ficial action, while others, perhaps because of differences in class back-
grounds or in the political culture of the institution, did not have offenses
recorded.
The example of the informal tracking of primary school children directs at-
tention to the second dimension of labeling by street-level bureau-
cracies-the implications of client status for the larger society. Not only do
slow-track children incorporate in their images of themselves the perspec-
tives of the teacher and the fast-track pupils. The label of"slow learners" also
has relevance for other people who play significant roles in the childrens'
lives. Parents learn that their children have been assigned to the slow track
and may begin to treat them as dull or with anxiety that they have been con-
signed early to failure. Brothers and sisters may tease. As Rist reports,
teachers in later years will treat the children as having the capacities as-
signed originally by the first teacher. There is little escape from the implica-
tions of this labeling. Sometimes people can alter stereotypical images of
themselves by presenting massive evidence to the contrary, 32 but in this
case the process of interaction is structured so that all the evidence seems to
confirm the original diagnosis.
Differences attributed to clients may be regarded as significant when peo-
ple important to the client respond to him or her differently. When dif-
ferences reach this level, a person's self-image and self-respect are affec-
ted.33 The status of "criminal," "juvenile delinquent," "welfare mother,"
and "slow learner" is stigmatic because it goes beyond mere distinctions
68
Relations with Clients
among people. Society takes these terms as signals to treat people dif-
ferently. As such, they are simultaneously cues to the people so labeled to
regard themselves differently.
Some differences assigned by street-level bureaucrats provide more sub-
tle stigma. For example, being a public housing tenant is not necessarily
stigmatic, but living in certain projects may be. Health treatment through
medicare may not affect a client's relationship with some clinics, but might
have substantial implications in others. The label of "troublemaker" may
predict official responses in some schools but not in others.
Prisoners are never fully destigmatized by the society after they have
served time. They become "ex-cons." Indeed, mere involvement with a bu-
reaucracy may be stigmatizing, even when the clearly stigmatic label is
avoided. This may be the fate of a defendant found innocent at trial, 34 or the
patient who is held for examination to determine whether he or she requires
treatment for mental illness, even if judged perfectly sane. If others begin to
treat one differently as a result of these labels, one may begin to incorporate
these views as his or her own self-image.
The particular difficulty with labels ascribed by street-level bureaucrats,
as we have seen, is that the characteristics on which they are based are prob-
lematic. Judgments concerning the status allocated by street-level bureau-
crats depend upon the discretion of the bureaucrat, which in turn depends
upon many indeterminate factors, such as training, the social context in
which the client is presented, and the presence or absence of similar "dif-
ferences" in the client population. The fast-track students in the above illus-
tration might have been labeled slow learners in white, middle-class schools
(the students and teachers were black). A criminal offense in one setting
might be overlooked in another. The social construction of the client, involv-
ing the client, others relevant to the client, and the public employees with
whom they must deal is a significant process of social definition often unre-
lated to objective factors and therefore open to the influences of prejudice,
stereotype, and ignorance as a basis for determinations.
Some bureaucracies so routinize their processing of clients that significant
psychological interactions are minimal. Welfare workers and legal service
lawyers, for example, may adhere to interview formats that exclude personal
elements and reduce the likelihood of decision making on the basis of inter-
personal interactions. 35 Of particular interest to those concerned with the
policy consequences of street-level bureaucrats' work is the ways in which
the tendencies to treat people in terms of their predicted behavioral charac-
teristics can be avoided. Some bureaucratic settings seem to result in stigma-
tic treatment and some do not. There are a variety of ways to interrupt track-
6g
CONDITIONS OF WORK
uals (for example, interviewing techniques), but virtually never how to jug-
gle case loads or handle large numbers of clients at one time, as initiates will
have to do. One of the best illustrations of the solidity of the myth of human
interaction in public services is provided by the transformation in the health
field of the word "care" from a verb to a noun. Politicians and administrators
regularly discuss levels and amounts of care that will be provided, but rarely
who will care, and how they will express their caring.
Advocacy
Street-level bureaucrats are often expected to be more than benign and pas-
sive gatekeepers. They are also expected to be advocates, that is, to use their
knowledge, skill, and position to secure for clients the best treatment or
position consistent with the constraints of the service. That street-level bu-
reaucrats should be advocates for clients is articulated explicitly in the pro-
fessional training and canons of lawyers, doctors, social workers, teachers,
and others. Those professions and semi-professions that display the altruism
critical to most definitions of professionalism require their members to make
clients' needs primary. Other street-level bureaucracies whose claims to
professional status are more questionable also display degrees of advocacy in
their obligations as public servants to be responsive to the citizens who pay
their salaries.
One source of the myth of service altruism is social policy reformers who
utilize the discrepancy between reality and stated policy intentions to mobi-
lize support for change. Virtually an policy reforms are advocated in the
name of achieving service ideals. (The myth of altruism does not assume
ideal policy implementation; it assumes only that policy and people who
implement it are well intentioned and that their work constitutes a net social
benefit.)
Perhaps even more important in sustaining the myth of service altruism
are the workers who attempt to implement the service ideal. Like most
social myths this one has a partial basis in reality. Each generation of workers
brings to its jobs, in addition to interest in material benefits, dedication to
helping people. Those who recruit themselves for public service work are at-
tracted to some degree by the prospect that their lives will gain meaning
through helping others.
The myth of human relationships in street-level bureaucracies no doubt
72
Advocacy and Alienation in Street-Level Work
partly accounts for whatever dedication exists to superior service and for
the ability of public services to realize their objectives. However, achieve-
ment of advocacy is undermined by several critical factors. Some of these
concerns were treated earlier in discussing the structure of street-level bu-
reaucrats' work, but it is useful to review briefly how the structure of work
and relations with clients compromise altruism and undermine advocacy
where advocacy is appropriate.
The helping orientation of street-level bureaucrats is incompatible with
their need to judge and control clients for bureaucratic purposes. This is evi-
dent in the following role tensions.
First, advocacy can only be done on behalf of single units, whether they
be individuals or collectivities such as a tenants' union. Moreover, the ad-
vocate must have enough free attention to devote to the client. This does not
mean that only one client can be dealt with at a time. But it does mean that
advocacy may be compromised by large case loads and mass processing of
clients. For the advocate, large case loads mean that every minute devoted
to one client means less time for others. Clearly organizations have to choose
what resources to provide, and a suboptimal amount is likely to be available
for any client. Street-level bureaucracies chronically tend to allocate rela-
tively low amounts of resources to facilitate workers taking clients' perspec-
tives. Clients are asked to understand this and even to incorporate this un-
derstanding into their concept of being a client. Those who fail to do so are
punished by bureaucracies, which can impose severe costs on clients who
contest the allocation of resources devoted to them.
Second, advocacy is incompatible with organizational perspectives. The
organization hoards resources; the advocate seeks their dispersal to clients.
The organization imposes tight control over resource dispersal if it can; the
advocate seeks to utilize loopholes and discret-ionary provisions to gain client
benefits. The organization seeks to treat all clients equally and to avoid hav-
ing to respond to claims that others received special treatment; the advocate
seeks to secure special treatment for individual clients. The organization acts
as if available resource categories had fixed limits (which is often not abso-
lutely true); the advocate acts as if resources were limitless (which is also not
true).
Street-level bureaucrats frequently encounter this tension. School coun-
selors are criticized for serving the interests of schools rather than individual
children. 2 Doctors have come under considerable criticism for failing to give
sufficient priority to not spending public monies for patients' health care.
This is another way of saying that their advocacy perspective (advocacy for
their patients and, perhaps, themselves) is not in balance with their cor-
73
CONDITIONS OF WORK
74
Advocacy and Alienation in Street-Level Work
child. In this four-sided relationship (agency, judge, parents, child) social
workers might well yearn for the clarity an entirely adversarial process might
provide. 8
Alienation
75
CONDITIONS OF WORK
strong. Street-level bureaucrats work in isolation, but they seek and receive
support from other workers. Street-level bureaucrats are no less peer-
related than other workers and find gratification in the squad room, teachers
lounge, and other places where they congregate.
However, there are several areas in which the alienation of street-level
work is fairly extensive. I have already said that the compromises required of
advocates reduce the extent to which street-level bureaucrats are able to
respond to clients in a fully human way. Moreover, street-level work is
inauthentic, in the particular sense of alienation delineated by sociologist
Amitai Etzioni. In its emphasis on providing services, mandating workers to
act as helpers, and giving them responsibility it "provides the appearance of
responsiveness while the underlying conditions . . . subject a person to
forces beyond his understanding and control." 11 In defense of the myth of al-
truism, street-level bureaucracies devote a relatively high proportion of
energies to concealing lack of service and generating appearances of
responsiveness.
In addition, street-level bureaucrats are alienated from clients-the prod-
uct of their work-in at least four particular respects: (1) they tend to work
only on segments of the product of their work; (2) they do not control the
outcome of their work; (3) they do not control the raw materials of their
work; and (4) they do not control the pace oftheirwork. 12 These considerations
are discussed below.
1. Working on segments of the product. One difference between shoe-
makers and shoe-factory workers is that the former crafts the entire shoe and
thus can draw satisfaction from seeing the fruits of his or her labor. The latter
may only cut the heel, punch holes for laces, glue parts together, or perform
some other aspect of the shoe making process. But he or she cannot take pride
in having made the article. This analysis of the implications of factory work is
as old as the factorization of production and as young as recent industrial ef-
forts to permit workers greater flexibility and variety in work assignments. 13
Street-level bureaucrats do not work on the entire product, but only on
segments. This is the case in two respects. In response to the need to cat-
egorize clients they tend to treat them only as bttndles of bureaucratically
relevant attributes rather than as whole persons. They deal with symptoms,
qualifications, and capacities, but not with feelings or superficially tangential
facts. The imperatives of processing people into the correct categories tend
to overwhelm both professional obligations to treat the whole person and the
recognition that responding to clients in narrowly defined areas is likely to
miss important dimensions of the presenting problem.
Advocacy and Alienation in Street-Level Work
Understanding that treating parts of people leads to inferior or inappro-
priate services has generally been the guiding idea behind many critiques of
social policy. For example, social workers and psychologists have been em-
ployed in hospitals, schools, and courts in efforts to respond holistically to
the citizen-client. Reformers recognize that a health problem such as lead
poisoning may be a problem of income and law. A problem of poverty may in
reality be a legal problem (e.g., being granted a divorce). Education prob-
lems may have emotional, physical, or economic origins. The more the soci-
ety recognizes the interconnectedness of service policy problems, and the
more the problems are or remain interconnected, the more the alienation of
categorization will impinge on street-level work.
Street-level bureaucrats also tend to work only on segments of the pro-
cess. In the name of efficiency, convenience, or optimal utilization of re-
sources the world of social services has become n:tore and more specialized.
Educators are math teachers, reading teachers, art specialists, dance thera-
pists. The world of special education now includes therapists on various
forms of reading disability, various degrees of mental retardation, and
various kinds of physical handicaps.
It would be one thing if these specialists had relatively intensive interac-
tions with children so that they could fully plan for them and have the time
to realize measurable results. But school programs tend to be filled with spe-
cialists who cannot take full responsibility for the product even if they
wanted to. This is not an argument against mainstreaming, for no doubt
children with special educational needs ought not to be segregated. But
mainstreaming in the modern school means going into a system that substan-
tially segments the child's school day. Schools not only specialize by func-
tion, but they also are organized so that different workers take responsi-
bilities for different stages through which pupils pass. Different teachers
receive students each year as teaching is cross-specialized by grade. The ex-
ample of specialization in teaching is particularly apt because schools have
the most extensive opportunities to interact substantially with service
recipients.
There are often considerable costs to specialization in addition to the
benefits of expertise and efficiency. Divisions between intake and casework
mean that interviews and fact gathering sometimes have to be repeated,
which is inefficient. When resource allocation does not permit extensive
reinterviewing of clients the result may be inappropriate decisions resulting
from the distortion of information between intake and case-work levels.
Recent proposals for multi-specialization in special education appear to pro-
77
CONDITIONS OF WORK
vide support for the view that clients of special education services have many
dimensions, and that treating the whole person is required ultimately for op-
timal results.
z. Controlling the outcome of the work. For reasons closely related to the
above, street-level bureaucrats often do not control the outcome of their
work. Specialization may mean that they do not see the work through or only
participate in a fraction of the work with clients. They do not control all of
the resources of the agency they work for. Sometimes they process people
for other bureaucracies, which ultimately disposes of cases. Police often
conplain, for example, that their actions are not supported by judges and
prosecutors; they feel uncertain whether their performance of a good job as
they define it will result in a desired outcome.
Another reason they do not control the outcome of their work is that
clients' problems are not subject to closure. Although street-level bureau-
crats are regarded, and regard themselves, as able to solve problems, the
problems do not end or are not resolvable. Many public service agencies are
called revolving doors for this reason. Street-level bureaucracies that are
oriented toward transforming clients, such as judicial institutions and social
welfare agencies, are revolving doors because the solutions they offer people
are not adequate. People do not stay "fixed." To the extent that this presents
street-level bureaucrats with severe dissonance between objectives and ca-
pabilities, they develop coping mechanisms to shield them from the implica-
tions of the gap between expectations and accomplishment. They are alien-
ated to the extent they experience this discrepancy as loss of control over
situations they are supposed to control.
3 Controlling the input. Street-level bureaucrats cannot control the na-
ture of the material with which they work. They cannot deploy to greatest ef-
fectiveness the skills they possess partly because the conditions of work pro-
hibit effective interaction with clients, and because they do not have control
over clients' circumstances even when conditions are favorable for interven-
tion. Even in institutions such as prisons and mental hospitals workers do
not control the under life of the institutions and cannot affect those realities of
clients' lives that contribute to their deviant behavior. How frustrating to be
a good teacher who has to greet every morning children who are hungry and
exhausted from lack of sleep. How frustrating to be a skilled professional at-
tempting to help a welfare recipient whose circumstances conspire to con-
found every constructive move. It takes a great deal of commitment or
cynicism to accept the futility of one's own efforts in such circumstances. 14
4 Controlling the work pace. Street-level bureaucrats do not control the
pace of work. I have said that their discretion provides a measure of reward
Advocacy and Alienation in Street-Level Work
on the job, but they often do not control the timing of their decision making.
This is the case in obviously reactive public service areas such as policing,
and in other areas where the amount of time spent on individual clients, or
the number of clients requiring attention, cannot be anticipated. Street-
level bureaucracies go to great lengths to produce predictability of client
demands, either by rationing services in some way or planning for peak
period workloads. To some extent they are able to develop devices to mini-
mize the costs of unpredictability. (This is treated in chapter 7.) But to some
extent they are not. Then they confront the problem that their work is their
master. They get behind in paperwork, are unable to respond to any but the
most minimal requests, and believe themselves to be ineffective because
they feel they have no slack time to respond fully to any individual situation.
Alienated work leads to dissatisfaction with the job. Job dissatisfaction af-
fects commitment to clients and to the agencies for which they work. The
proposition that street-level bureaucrats perform in alienated labor roles
contributes to understanding the dynamics of some recent developments in
public service organization.
Implications of Alienation
79
CONDITIONS OF WORK
ther. Thus the working conditions that give rise to alienation in work may
cumulatively contribute to separating the client from the public service
worker. This is significant since in earlier periods public service workers
have often championed client rights and benefits. There have been times in
American labor history when fledgling public service unions and worker as-
sociations bargained for clients as well as for themselves (social workers and
teachers, for example). This struggle has become less important as the con-
nection between workers and clients has dissipated.
In general, employers confronted by alienated workers can choose among
the following responses in some combination. They can tgnore the situation
and accept the absenteeism, low morale, poor performance, and other mani-
festations of worker dissatisfaction. They can restructure the work to make it
less alienating. Or they can concentrate on changing the mix of benefits and
sanctions that they offer workers outside of working conditions. They can
increase pay for work that is less rewarding or improve conditions tangential
to the work. Or they can raise the costs of nonproductive behavior.
The trends in public employment since, say, 1g6o reflect these alterna-
tives. Public service workers have increased their share of national wealth
through higher pay and benefit levels, increased their collective bargaining
power, and acquiesced in and often encouraged developments such as spe-
cialization, computerization, and fragmentation of responsibilities for clients.
Street-level bureaucrats have enhanced their position in the political system
to the neglect of aspects of service consistent with more humanistic models
of client involvement, or at the expense of taking positions on clients' behalf.
These statements no doubt require qualification for particular service areas,
but at a general level they suggest linkages between the quality of client ser-
vices, the structure of street-level bureaucrats' work, and the priorities of a
society oriented toward cost effectiveness and the dispensary model of pro-
viding for human needs.
8o
PART III
PATTERNS OF
PRACTICE
Introduction
they are doing a perfect job or performing the way the job should be per-
formed; only that they are functioning effectively and properly under the
constraints they encounter. The typical teacher, policeman, welfare
worker-ind eed anyone who regularly meets the public-seem s to have an
image of himself or herself as working under great strain and with consider-
able sacrifice to provide clients protection or service no one else would be
willing to provide. They see themselves as fighting on the front line of local
conflict with little support and less appreciation by a general public whose
dirty work they do. 3
If they have any recognition that their performance is less than adequate
under the circumstance s they confront, they are likely to seek and find the
explanation someplace other than in their own inadequacy. Street-level bu-
reaucrats who are unable to retain a concept of their own adequacy in the job
are more likely to leave it or seek other work than to sustain the personal am-
bivalence that results. This orientation applies even though the civil service
system may deaden motivation, and some public employees may be mo-
tivated by primarily selfish consideratio ns.
The point that they try to do a good job in some way fulfills the require-
ments for asserting that the problem of street-level bureaucrats is one of
decision making under conditions of considerable uncertainty where satisfac-
tory decisions about resource allocation must be personally as well as organi-
zationally derived. The work context of street-level bureaucrats calls for the
developmen t of mechanisms to provide satisfactory services in a context
where the quality, quantity, and specific objectives of service remain (within
broad limits) to be defined.
There is by now a venerable tradition in organization al studies concerning
the search for satisfactory rather than optimal solutions to decision-ma king
problems under conditions of uncertainty. 4 But the analysis of street-level
bureaucracy may be somewhat different from other studies because it is not
only the decisions that become satisfactory rather than optimal, but also the
mental and organization al processes that must become satisfactory. Thus to
understand street-level bureaucracy one must study the routines and subjec-
tive responses street-level bureaucrats develop in order to cope with the dif-
ficulties and ambiguities of their jobs. 5
We can now restate the problem of street-level bureaucracy as follows.
Street-level bureaucrats attempt to do a good job in some way. The job,
however, is in a sense impossible to do in ideal terms. How is the job to be
accomplishe d with inadequate resources, few controls, indetermina te objec-
tives, and discouraging circumstanc es?
There are three general responses that street-level bureaucrats develop to
Introduction
deal with this indeterminacy. First, they develop patterns of practice that
tend to limit demand, maximize the utilization of available resources, and
obtain client compliance over and above the procedures developed by their
agencies. They organize their work to derive a solution within the resource
constraints they encounter. Second, they modify their concept of their jobs,
so as to lower or otherwise restrict their objectives and thus reduce the gap
between available resources and achieving objectives. Third, they modify
their concept of the raw materials with which they work-their clients-so
as to make more acceptable the gap between accomplishments and objec-
tives. Much of the patterned behavior of street-level bureaucrats, and many
of their characteristic subjective orientations, may be understood as re-
sponses to the street-level bureaucracy problem.
which the confronting problem is the management of work stresses. The ten-
dency to seek control over the work environment is perhaps evident in most
work situations, but again it is worthy of comment because in street-level bu-
reaucracies the search is typically a matter of public policy. Routines could
be structured to maximize the achievement of agency objectives. Or they
could be structured to maximize responsiveness to clients. No doubt these
competing perspectives do account for workers' routines to some degree.
However, the extent to which routines are structured to maximize worker
control over the work context may measure the extent to which articulated
agency policy objectives are difficult to achieve.
The routines of work in street-level bureaucracies appear to be directed
toward achieving one or more of four purposes in processing clients.
1. They ration services.
2. They control clients and reduce the consequences of uncertainty.
3 They husband worker resources.
4 They manage the consequences of routine practice.
86
CHAPTER 7
Theoretically there is no limit to the demand for free public goods. Agencies
that provide public goods must and will devise ways to ration them. To ra-
tion goods or services is to establish the level or proportions of their distribu-
tion. This may be done by fixing the amount or level of goods and services in
relation to other goods and services. Or it may be done by allocating a fixed
level or amount of goods and services among different classes of recipients.
In other words, services may be rationed by varying the total amount avail-
able, or by varying the distribution of a fixed amount.
This usage is consistent with the familiar application of rationing in war-
time. During World War II, for example, automobile tires were rationed by
restricting their production for domestic purposes and limiting individual
purchases, making them costly, and establishing priorities among users (doc-
tors were privileged in this respect). This chapter considers rationing in
street-level bureaucracies that has the effect of fixing (usually to reduce or
limit) the level of services. The next chapter takes up rationing that differen-
tiates among clients.
The rationing of the level of services starts when clients present them-
selves to the worker or agency or an encounter is commanded. Like factory
workers confronted with production quotas, street-level bureaucrats attempt
to organize their work to facilitate work tasks or liberate as much time as pos-
sible for their own purposes. This is evident even in those services areas in
which workers have little control over work flow. For example, police often
cannot control work flow because most police assignments are in response to
PATTERNS OF PRACTICE
88
Rationing Services: Limitation of Access and Demand
those who do not. When people have them they enhance personal influence.
When workers for public agencies have them they may be used to direct or
subordinate clients or discourage clients from further interactions with the
agency.
MONETARY
Street-level bureaucracies can rarely assign monetary costs for services,
since by definition public services are free. However, monetary costs are
imposed in several instructive instances. In income-providing programs citi-
zens' contributions to the income package may be manipulated as policy.
Medicare patients may be asked to pay a higher deductible before insurance
provisions become operable. Food-stamp recipients may be asked to pay
more for their stamps. The effective taxation of earned income in welfare
reduces the number of people in contact with this street-level bureaucracy.
Clearly differences in monetary costs serve to ration street-level bureaucrats'
services.
Programs sometimes force clients to incur monetary costs that discourage
them from seeking service. Acquiring records from other agencies to es-
tablish eligibility or securing transcripts for appeals can be costly, particu-
larly if travel is involved. Agencies that keep bankers' hours impose mone-
tary costs on working people who cannot appear without losing wages.
Appointments sometimes require parents to seek babysitters. Street-level
bureaucracies that seek to minimize these penalties introduce evening office
hours, or they provide child-care services.
TIME
Just as available time is a resource for people in politics, it is also
a unit of value that may be extracted from clients as a cost of service. Clients
are typically required to wait for services; it is a sign of their dependence and
relative powerlessness that the costs of matching servers with the served are
home almost entirely by clients. It is to maximize the efficiency of workers'
time that queues are generally established. A primary reason that clinic-
based practice is more efficient than home-based practice is simply that it is
patients and not physicians who spend time traveling and waiting. Police-
men also allocate time costs by stopping to question young people who,
while not guilty of any crime, are judged to require reprimanding. 4
Some teachers in some school systems make home visits to meet with
parents, while others schedule parent-teacher conferences after school on
specific days set aside for such purposes. (If there are two parents and one or
both work, both are unlikely to be able to meet with the teacher.) These al-
8g
PATTERNS OF PRACTICE
INFORMATION
Giving or withholding information is another way in which services
may be rationed. Clients experience the giving or withholding of infor-
mation in two ways. They experience the favoritism of street-level bureau-
crats who provide some clients with privileged information, permitting
them to manipulate the system better than others. And they experience it
as confusing jargon, elaborate procedures, and arcane practices that act as
barriers to understanding how to operate effectively within the system. The
emblematic carrier of this characteristic is the court clerk who runs his words
together in an undecipherable litany to the dominance of court procedures
over citizens' rights. 7 At the bureaucratic level the giving and withholding of
information is most obvious in examining how agencies manipulate their case
loads by distributing or failing to distribute information about services.
Conventionally, analysts assess the demand for services by studying client
rolls and visits. (Demands are statements directed toward public officials
that some kind of action ought to be undertaken.) 8 If it is recognized that
manifestations of client involvement may not fully reflect client interests,
go
Rationing Services: Limitation of Access and Demand
analysts contrive ways to assess underlying needs, for example, through atti-
tudinal and census surveys. From this assessment administrators and politi-
cians make claims about appropriate levels of services.
However, if it is recognized that organizations normally ration services by
manipulating the nature and quantity of the information made available
about services, then it is easily seen that demand levels are themselves a
function of public policy. Client rolls will be seen as a function of clients' per-
ception of service availability and the costs of seeking services. Client de-
mand will be expressed only to the extent that clients themselves are aware
that they have a social condition that can, should, and will be ministered to
by public agencies.
When New York City reduced acceptance rates for new welfare cases at
seven centers by 17 percent it accomplished this feat by tightening the
application process. This meant not only more careful scrutiny of applicants'
claims, but also more documentation and inquisition was required, which
contributed a separate measure of rationing. 9
This perspective is illustrated by indices of need for legal assistance for
domestic problems. When a sample of Detroit residents were asked if they
required a lawyer for assistance with some domestic-relations matters,
scarcely more than 1 percent answered affirmatively. It would have been dif-
ficult to predict from this survey that approximately 40 percent of the clients
of legal aid and neighborhood law offices originally sought help with domes-
tic problems. 10
Needs become manifest when the institutions that might provide assis-
tance send out signals that they stand ready to assist. The 40 percent of the
clients who originally sought help with domestic matters might have been
only a small portion of the population that could have benefitted from such
assistance. Some who could have used such services may have been deterred
from seeking them. Since legal services are vastly underfunded, even more
dramatic demonstrations of need might have materialized if more lawyers
had been available.
Information about service is an aspect of service. Withholding information
depresses service demands. For example, the campaign to reform welfare by
dramatically increasing the welfare rolls was based on the view that a politi-
cal movement could help overcome the stigma attached by potential recipi-
ents to welfare status. It could provide the information necessary to realize a
substantial increase in the number of recipients. 11 The failure of public wel-
fare agencies to make sure potential recipients receive the benefits to which
they are entitled contrasts dramatically with the success of social security
PATTERNS OF PRACTICE
92
Rationing Services: Limitation of Access and Demand
ward making the agencies look better. For example, when drug treatment
centers were few they could afford to impose rigorous residential require-
ments, particularly since clients' commitment to their own rehabilitation was
considered critical to therapy. When the number of such institutions in-
creased in the early 1970s in response to available funding, and the popula-
tion of drug users started to decline, to increase their clientele the centers
began to relax their enrollment requirements (for example, by accepting
clients who previously would have been judged too difficult to help). They
also relaxed attendance requirements, so that a treatment bed might be oc-
cupied by someone who was not in fact a full-time resident of the center. Be-
sides drug treatment centers, other organizations that have competed for
larger shares of a fixed client pool include mental health centers funded in
the same city, and academic departments competing for students within a
university.
In theory this bureaucratic competition might provide precisely what bu-
reaucracies importantly lack-a substitute for market place accountability.
This, of course, is the idea behind educational vouchers. However, the
healing effects of competition are too often mitigated by the residual bureau-
cratic aspects of the competing organizations. Faculty members in academic
departments with declining enrollments are still protected by the tenure
system, rewards for research (and bringing in research grants), and other fac-
tors that protect them from being assessed solely on criteria of service to
students. Similarly, educational voucher experiments have foundered on
teachers' tenure, union opposition, and parental inability to express prefer-
ences within the system for lack of information on the implications of the
available choices.
PSYCHOLOGICAL
Bureaucratic rationing is also achieved by imposing psychological costs on
clients. Some of these are implicit in the rationing mechanisms already men-
tioned. Waiting to receive services, particularly when clients conclude that
the wait is inordinate and reflects lack of respect, contributes to diminishing
client demands. 16 The administration of public welfare has been notorious
for the psychological burdens clients have to bear. These include the degra-
dation implicit in inquiries into sexual behavior, childbearing preferences,
childrearing practices, friendship patterns, and persistent assumptions of
fraud and dishonesty. 17 Nor have these practices been confined to the "un-
enlightened" 1950s, although some of the more barbaric features of welfare
practice, such as the early dawn raids to catch the elusive "man-in-the-
house," are no longer practiced.
93
PATTERNS OF PRACTICE
94
Rationing Services: Limitation of Access and Demand
Queuing
The most modest arrangements for client servicing impose costs on clients.
This is evident in the way clients are arranged, or required to present them-
selves, for bureaucratic processing. Even the most ordinary queuing ar-
rangements--th ose designed to provide service on a first-come, first-served
basis in accordance with universalistic principles of client treatment-im-
pose costs. 20
Queues that depend upon first-come, first-served as their organizing prin-
ciple elicit client cooperation because of their apparent fairness, but they
may ration service by forcing clients to wait. When clients are forced to wait
they are implicitly asked to accept the assumptions of rationing: that the
costs they are bearing are necessary because the resources of the agency are
fixed. They are also controlled by the social pressures exerted by others who
wait. This is one of the functions of the line, waiting room, and other social
structures that make it evident that others share the burden of waiting for
service.
While resource limitations may be unalterable in the very short run, they
are not necessarily immutable. They derive from allocation decisions that
consider it acceptable to impose costs on waiting clients. Costs will not be
imposed upon clients equally. Long lines processed on a first-come, first-
served basis relatively benefit people who can afford to wait, people whose
time is not particularly valuable to them, or people who do not have other
obligations.
Poor people often suffer in such a system. Not only may clients who ap-
pear more affiuent get served first because it is thought that the costs of wait-
ing are higher for them, 21 but agencies often paternalistically develop policy
as if the costs to the poor were nonexistent. A visit to the waiting room of a
welfare office in any inner-city neighborhood is likely to convey the impres-
sion that the Welfare Department assumes recipients have nothing else to
do with their time. Recipients learn the lesson of people who must seek ser-
vice from a single source. Like the telephone company, the welfare depart-
ment is able to pass on to the customer the costs of linking people with ser-
vice. This system also benefits the average client to the disadvantage of
people with extraordinary needs, since initially it has no mechanism for dif-
ferentiating among clients. However, where the injury to people with ex-
traordinary needs is likely to be severe, as in police work or medical
emergencies, the ordering of services is often deliberately structured to
search for and respond to this information.
95
PATTERNS OF PRACTICE
Similar to the queue by appointment is the waiting list; clients are asked to
wait for what is usually an undetermined amount of time until they can be
accommodated. Although it appears to be straightforward on the surface, the
g6
Rationing Services: Limitation of Access and Demand
waiting-list system has several important latent functions. First, as we have
seen in the case of Boston public housing, a waiting list tends to increase the
discretion of street-level bureaucrats by providing opportunities to call
clients from the waiting list out of tum, or to provide special information that
will permit them to take advantage of ways to be treated with higher prior-
ity.24 Waiting lists also permit agencies to give the appearance of service
(after all, clients are on a waiting list) and to make a case for increased
resources because of the backlog of demand. 25 The waiting list appears to
record the names of potential clients who are seeking service but cannot be
accommodated, although it is obvious to all that many names continue on the
list only because the agency has not attempted to discover who is actively
waiting and who has long since ceased to be interested.
Some social agencies act as if the waiting list usefully filters potential
clients who are truly in need of service and strains out those whose needs are
not substantial and who thus drop off. This system of rationing may also pro-
vide for a period of time in which spontaneous recuperation may occur, again
reserving client spaces only for those who are needy. 26 However, it is uncer-
tain whether continuation on the list is a sign of substantial need or precisely
the opposite, a sign that the potential client is successful enough in managing
the problem that he or sl{e can wait patiently for services.
A queuing arrangement that maximizes the costs to citizens at the expense
of a relatively small number of street-level bureaucrats is employed by lower
courts, which typically require defendants to appear on a given day, but no-
tify them only as to the hour they should appear. In a typical situation fifty to
one hundred defendants, possibly with a friend or member of their family,
must be ready for a hearing or arraignment, with substantial penalties if they
do not appear precisely at the beginning of the session (when their names are
first called). Here they must wait until the judge arrives, and then wait again
while the judge gives priority to defendants in the lockup who may require
attorneys, defendants whose attorneys plead that they have to be elsewhere,
and defendants whose cases require the testimony of waiting police officers,
who themselves are subject to other priorities. Only when these and other
priorities are accommodated will the docket be called in alphabetical or
some other order.
Defendants may be innocent but by virtue of being arrested are judged
guilty enough to pay in time and uncertainty the price that the court exacts
for scheduling cases for the primary convenience of the judge. Although
practices vary from court to court it is typical that defendants will not be told
even approximately when their cases will be called, so that they must wait in
the courtroom, possibly for most of the day, until they receive a hearing. 27
97
PATTERNS OF PRACTICE
The defendant who has waited through such a day has been instructed in the
costs of continued interaction with the court system and must consider
whether exercising rights or even pleading innocent in a minor matter, al-
though legally valid, is worth the time and irritation. Some court systems
have recently recognized that similar problems, including frequent post-
ponements, inhibit witnesses from appearing and testifying in trials. But the
same analysis rarely focuses on defendants and their experiences in court.
This queue by roundup is also typical of jury impaneling, where citizens
are called for a week of service and must sit in a jury room awaiting assign-
ment, often for several days, perhaps never to be called. The system of-
ficially is justified by the fluctuating and relatively unpredictable demand for
jurors, and again is premised on the high value placed on the court's time
relative to citizens' time. To insure that there are always people ready to
serve, more jurors are called than will be required. If the court could toler-
ate a postponement now and then for lack of available jurors, and if jurors
were called to report serially during the week rather than all at once, less
time would be wasted for prospective jurors. But such practices could only
be adopted if the time of prospective jurors were accorded more value rela-
tive to judges' and lawyers' time than is currently the case.
Clients frequently may be quite willing to pay the costs of waiting. Clients
undoubtedly understand that there are times when they will have to wait,
unless bureaucracies hire enough staff to meet peak demand. And since
demand in most street-level bureaucracies is to some degree unpre-
dictable--even schools often have to hire new teachers or shuffie teacher as-
signments after school has started-it would be too costly to provide services
so that waiting would never occur. Waiting becomes injurious and inappro-
priately costly only under certain conditions.
Waiting is inappropriate when it exceeds the time generally expected for a
service. A person may not resent a two-hour wait in an emergency room to
receive a tetanus shot if it is clear that patients with more serious claims are
being served first. But the same amount of time spent waiting in line simply
to hand in forms to renew a driver's license may be exceedingly irritating.
Waiting may also be resented when it involves the violation of an implicit
agreement. Waiting is regarded as inappropriate when clients have made an
appointment, except when the appointment is considered only an approxi-
mation of the time of service (as in the case of office visits to doctors).
Still another situation in which clients resent the costs of waiting arises
when they wait unfairly. Thus if a favored client gains access to service more
easily than others it will be resented by those who are not favored. Some-
times unfairness in waiting time may be so slight as to go unnoticed by
g8
Rationing Services: Limitation of Access and Demand
clients. A study of black patients in Chicago hospital emergency rooms re-
vealed that compared to whites waiting time was a little more than three
minutes, incurred primarily by claimants with nonemergency conditions
who sought help when the emergency room was relatively busy. But this cost
is not actually trivial. It is worth noting that a modest three minutes or so,for
the 1 , 1 os blacks in the sample alone, would add up to a full working day for
2,619 people on a yearly basis, 28 a measure of one of the costs of institutional
racism for the blacks of Cook County, Illinois.
The existential problem for street-level bureaucrats is that with any single
client they probably could interact flexibly and responsively. But if they did
this with too many clients their capacity to respond flexibly would disappear.
One might think of each client as, in a sense, seeking to be the one or among
the few for whom an exception is made, a favor done, an indiscretion over-
looked, a regulation ignored.
This dilemma of street-level bureaucrats is illustrated well by the legal
services program. Individually, each attorney is obliged by professional
norms to pursue fully the legal recourses available to clients. For impover-
ished clients this presumably means that attorneys should act on clients'
behalf irrespective of cost. Only if this assumption is correct could the provi-
sion of legal services begin to redress the balance of power in the legal sys-
tem, which every observer concedes favors those who command legal re-
sources. But if all clients' legal needs were fully pursued there would be no
time for additional clients. The dilemma is exquisite. To limit lawyers' ad-
vocacy is to deny poor people equal access to the law. To permit unbounded
advocacy is to limit the number of poor people who can have such access.
Only a reconstitution of the legal system could overcome the dilemma within
the current patterns of inequality: either a radical departure in the amount of
subsidies for legal assistance for the poor or a radical simplification of legal
procedures.
When confronted with the dilemma of serving more clients or maintaining
high quality service, most public managers will experience great pressures
to choose in favor of greater numbers at the expense of quality. Their inabil-
ity to measure and demonstrate the value of a service, when combined with
high demand and budgetary concerns, will tend to impose a logic of increas-
99
PATTERNS OF PRACTICE
ing the quantity of services at the expense of the degree of attention workers
can give to individual clients. Street-level bureaucrats, however, may devise
ways to sabotage management efforts to reduce interactions with clients. The
costs of achieving compliance in the face of workers' resistance may some-
times be more than managers want to pay. An example of such worker resis-
tance is related by Robert Perlman in his study of the Roxbury Multi-Service
Center.
Confronted with the complexity and number of demands being made on them, staff
members resorted to shielding themselves fiom the mounting pressures. They ex-
tended interviews to postpone or avoid taking the next client. They scheduled home
visits in order to avoid intake duty. 29
Whether street-level bureaucrats oppose efforts to limit their interaction
with clients, or whether they accept and encourage such efforts as a way of
salvaging an unattractive or deteriorating work situation, is perhaps the criti-
cal question on which the quality ofpublic service ultimately depends. Al-
though street-level bureaucrats may sometimes struggle to maintain their
ability to treat clients individually, the pressures more often operate in the
opposite direction. Street-level practice often reduces the demand for ser-
vices through rationing. The familiar complaints of encountering "red tape,"
"being given the run-around," and "talking to a brick wall" are reminders
that clients recognize the extent to which bureaucratic unresponsiveness pe-
nalizes them.
Routinization rations services in at least two ways. First, set procedures
designed to insure regularity, accountability, and fairness also protect work-
ers from client demands for responsiveness. They insulate workers from hav-
ing to deal with the human dimensions of presenting situations. They do this
partly by creating procedures to which workers defer, happily or unhappily.
Lawyers and judges, for example, generally accept court procedures that
insulate them from erratic client demands. Police officials resist instituting
(or more properly, reinstituting) a beat system because they are apprehen-
sive that officers would become too involved with neighborhood residents,
and thus perhaps engage in biased behavior. For similar reasons they often
oppose assigning officers to the areas in which they reside, and they advocate
reasonably frequent changes in assignment.
Social workers may be unhappy with the requirement to process endless
paperwork rather than spend time providing client services. But whether
happy or unhappy with job routines the fact remains that they serve to limit
client demands on the system. The righteous objections of critics that rou-
tine procedures detract from primary obligations to serve clients are of little
100
Rationing Services: Limitation of Access and Demand
account, since in an important sense it is not useful for the bureaucracies to
be more responsive and to secure more clients.
Second, routines provide a legitimate excuse for not dealing flexibly, since
fairness in a limited sense demands equal treatment. Unresponsiveness and
inflexibility reinforce common beliefs already present that bureaucracy is
part of the problem rather than the solution, and they further reduce clients'
claims for service or assertions of need.
When routines lead to predictability they may promote a degree of client
confidence. As a public defender lecturing his peers on increasing client
trust advised: "It's better to tell a client you will see him in two weeks and
then show up, than to reassure him by saying, Til stop by tomorrow,' and
never show. "30
But agency practices do not always lead to predictability. When they lead
to delay, confusion, and uncertainty they assign considerable costs to clients.
At times routines established to protect clients are distorted to minimize
contact or services. For example, to insure responsiveness housing inspec-
tors may be required to make more than one effort to contact complainants.
However, inspectors may become adept at telephoning complainants when
they are unlikely to be home or fail to keep appointments punctually. In Bos-
ton this practice "enhanced the prospects of no one being horne when the
inspector arrived-a practice which when repeated thrice, enabled cases to
be dropped. " 31
The significance of practices that subvert predictability, antagonize or ne-
glect clients, or sow confusion and uncertainty is that they are generally
functional for the agency. They limit client demands and the number of
clients in a context where the agency has no dearth of responsibilities and
would not in any way be harmed as an agency if clio:nts became disaffected,
passive, or refused to articulate demands. Any redr ction in client demand is
only absorbed by other clients who come forward, c r by a marginal and insig-
nificant increase in the capacity of street-level bmeaucrats to be responsive
to the clients who continue to press.
It is for this reason that we conclude that stated intentions of street-level
bureaucracies to become more client-oriented, to receive more citizen
input, and t.:> encourage clients to speak out are often questionable, no mat-
ter how sincere the administrators who articulate these fine goals. It is dys-
functional to most street-level bureaucracies to become more responsive.
Increases in client demands at one point will only lead to mechanisms
to ration services further at another point, assuming sources remain
unchanged.
101
PATTE RNS OF PRACT ICE
The logical but absurd extension of the relationshi p between demand and
services is exemplified by the apocrypha l library that reduced costs by clos-
ing down. Yet it is a real problem that increased patronage of libraries, mu-
seums, zoos, and other agencies providing free goods increases their uncom-
pensated costs when they succeed in becoming more attractive.
Undoubted ly there are dimension s of bureaucrat ic practice in which in-
creased responsive ness does not add to workers' tasks. Addressin g clients
politely rather than rudely or indifferent ly is an area in which greater re-
sponsivene ss is not necessarily burdensom e to the work load. Furthermo re,
reorganiza tion may result in increasing the responsive capacity of workers.
However, most increases in responsive ness--doin g more for clients, or even
listening to them more-pla ce additional burdens on street-leve l bureau-
crats, who will subvert such developme nts in the likely absence of any strong
rewards or sanctions for going along with them.
There are times when bureaucra tic rationing is not simply implicit; limit-
ing clientele or reducing services is the agency's stated policy. In response to
reduced budgets or other developme nts that make client-wor ker ratios con-
spicuously high, agencies will reduce the scope of service in several charac-
teristic ways. In reducing services explicitly they will continue to honor the
formal norm of universalistic service patterns.
Street-leve l bureaucrac ies may reduce services geographically. They may
formally narrow the catchment area from which clients are drawn or reduce
the number of neighborh oods served by a program. Alternatively, because
reductions in service are unpopular , street-leve l bureaucrac ies may prefer to
reduce the number of centers, effectively cutting services to some areas
without formally changing anyone's eligibility. When the borough of Man-
hattan, for example, consolidat ed its municipal court system, eliminatin g
district courts in Harlem, it did not formally change access to the court, but
informally it substantial ly increased the costs of using the court system to
Upper Manhattan residents.
Services can be limited in terms of clients' personal characteris tics. For-
mally, agencies can change income eligibility levels. Informally, they may
limit service by failing to print posters in Spanish or by placing notices in
old-age and nursing homes rather than in public housing in order to attract
primarily an elderly population .
Street-leve l bureaucrac ies also can formally or informally ration services
by refusing to take certain kinds of cases. The decriminal ization of drunken-
ness, for example, formally exonerates policemen from dealing with alco-
holics (although public disapproval still places pressure on the police to do
something about drunks). Informally departmen ts can limit the clientele if
102
Rationing Services: Limitation of Access and Demand
officers choose to ignore public drunkenness, or they can reduce its place in
departmental priorities.
Even when limiting services is not explicitly the function of rationing prac-
tices, service limitation often is not an unintended consequence of bureau-
cratic organization. Street-level bureaucrats and agency managers are often
quite aware of the rationing implications of decisions about shorter office
hours, consolidation of services, more or fewer intake workers, or the avail-
ability of information. Consider, for example, the efforts of the Budget
Bureau of New York City in 1g6g to decrease welfare expenditures. In a doc-
ument remarkable for our purposes the Bureau suggested several ways to
save close to $100 million. 32 In addition to reducing allowance levels, which
would supply the bulk of the savings, the bureau recommended four ad-
ministrative changes. Each would explicitly ration services in some way. A
new intake procedure was proposed that would require applicants to be ac-
tively seeking jobs prior to the intake interview. This would force people to
accept low-wage work, and, it was hoped, "more aggressive utilization of ex-
isting leverage over the employables would ... have a deterrent effect on
applications for welfare. "33 The authors recognized that for this innovation to
be effective a substantially greater capacity of public employment agencies
would be required, but there was no discussion of the costs of achieving this
increase.
More frequent recertifications would be conducted to induce recipients
who were on the rolls but no longer eligible because of changed circum-
stances to initiate case closings. (More than half of all case closings were then
initiated by clients.) This reform would reduce the time between changes in
clients' circumstances and the next reporting period.
Closing seven outreach centers would save some of the costs of running
the centers, but more importantly, "larger savings are anticipated from sec-
ondary effects .... The most important of these is the opportunity to build
up and maintain the maximum legal backlog between intake and eligibility
increasing average backlog from two weeks to a full month." 34 Among other
secondary benefits of center closings, the authors of the recommendations
expected that "the relative inconvenience to the client of self-maintenance on
emergency grants (for which application is normally made at the center more
than once a week) may have some deterrent effect on [those] marginally
eligible for welfare." 35
Finally, stronger management audits would introduce greater uniformity
in the system and provide better checks on welfare employees, who are por-
trayed in the document as more interested in enrolling clients than in con-
trolling welfare costs.
103
PATTERN S OF PRACTIC E
104
CHAPTER 8
Rationing Services:
Inequality in Administration
Free public goods and services may be rationed by imposing costs and fixing
their amount. They may also be rationed by allocating them differentially
among classes of claimants. In street-level bureaucracies services are distrib-
uted differentially for at least four interconnected reasons.
First, as mentioned before, to a degree the society wants bureaucracies to
be capable of responding flexibly to unique situations and to be able to treat
people in terms of their individual circumstances. This is particularly the
case for street-level bureaucracies. Teachers are expected to be interested in
the individual child, policemen to he capable of flexible responses, social
workers to be attuned to individual needs.
Second, street-level bureaucrats often want to make an improvement in
their clients' lives. They derive satisfaction from making a difference for
some clients and resist efforts to reduce the discretion that permits them to
have this influence.
Third, and most obviously, bureaucracies are simply often required to dif-
ferentiate among recipients. Everyone is not equally entitled to public ser-
vices. Eligibility, culpability, and suitability for bureaucratic intervention
must all be determined. Indeed, the process of reducing a person to his or
her qualifications for bureaucratic intervention essentially is the process of
becoming a client. 1 In this respect people-processing bureaucracies have
two tasks: to develop an appropriate set of categories in terms of which peo-
ple will be processed, and to map clients in terms of their qualifYing or dis-
qualifYing characteristics.
105
PATTER NS OF PRACT ICE
106
Rationing Services: Inequality in Administration
the ambiguities and psychological stresses of their jobs. Client differentia-
tion may take place because, confronted with heavy work loads and ap-
parently impossible tasks, street-level bureaucrats seek ways to maximize
personal or agency resources, or they attempt to succeed with some clients
when they cannot succeed with all. Sanctioned bureaucratic differentiation,
for which triage is the paradigm, is open to the potential that street-level bu-
reaucrats will differentiate among clients for reasons having more to do with
solving or resolving work-related problems than with providing optimal re-
source distribution. In this they commonly introduce precisely the sort of
particularism that modem bureaucracy theoretically overcomes, however
firmly the rules of the agency require universalistic standards of judgment.
I have said that the routines of street-level bureaucracies may well be the
policies of the agency, whether or not they are consistent with agency regu-
lations and standards. To lend substance to this observation some of the pat-
terns of practice by which discretionary judgments in these organizations are
made should be identified.
CREAMING
Confronted with more clients than can readily be accommodated street-
level bureaucrats often choose (or skim off the top) those who seem most
likely to succeed in terms of bureaucratic success criteria. This will happen
despite formal requirements to provide clients with equal chances for ser-
vice, and even in the face of policies designed to favor clients with relatively
poor probabilities of success. Employment counselors, for example, may
send to jobs people who have the greatest chance to gain employment any-
way, to the neglect of people who are more difficult to place. The Upward
Bound program, dedicated to enriching the educational backgrounds of dis-
advantaged high school students, constantly had to guard against projects tak-
ing students whose chances of getting into college were already fairly high. 4
Why does creaming take place, particularly in the face of official opposi-
tion to the practice? In every case of creaming the agency's incentives re-
ward successes with clients, but they provide no substantial rewards for the
risks taken. Since not all potential clients can be served, the reward struc-
ture of the agency is adopted as its implicit agenda in the absence of powertul
incentives to the contrary. If all clients are equally worthy but all cannot be
served, increasing the rate of personal or agency success becomes primary.
The situation is complicated by the fact that the criteria for determining who
is or is not "high risk" or "likely to succeed" are so problematic that there are
few clearcut ways to challenge the worker who may offend selection norms.
From another point of view creaming takes place in circumstances in
PATTER NS OF PRACT ICE
WORKER BIAS
Differentiation among clients may take place because of workers' prefer-
ences for some clients over others. They may prefer some clients over
others, despite official norms to the centrary, under at least three circum-
stances. Ultimately they all reflect the fact that workers find greater
gratification in interacting with some clients than with others and have op-
portunities to act on these preferences.
First, some clients simply evoke workers' sympathy or hostility. Like the
Israeli customs officials, workers may be inclined to "give the underdog a
break" 5 or may favor clients with similar ethnic backgrounds , as when racial
or ethnic favoritism prevails in discriminato ry decision making. The Boston
Housing Authority workers who tended to favor white elderly applicants
probably were responding to both ethnic and sympathy appeals when they
selectively provided them with critical information.
It would be as much of a mistake to infer that ethnic or racial appeals
always prevail in affecting discretionary judgments as that they never pre-
vail. Bureaucratic norms operate to restrict the range of determinatio ns
made in this way. Thus, black police officers may make particular efforts to
act in role-prescrib ed ways when confronting black citizens. 6 Displaying the
complement ary tendency, white bureaucrats may be more lenient or toler-
ant with black clients out of fear of being accused of racial biases. The report
from San Francisco that black school children tended to receive good grades
and were told that they were doing well in school, but in fact were not learn-
ing at an acceptable rate, is a vicious example of what can happen when
street-level bureaucrats over-react to the potential for biased behavior.
If public officials were simply biased or racist, and if their prejudices were
regularly manifested in behavior, the problem of bias in bureaucracy would
be more pernicious but easier to root out. At the very least it would be easier
to establish policy directives to reduce bias in bureaucracy . But patterns of
108
Rationing Services: Inequality in Administration
prejudice are more subtle in the modern bureaucracy dedicated officially to
equal treatment. Modern bureaucracy promises to eradicate prejudicial be-
havior through universalistic treatment; when prejudice does occur, it is
more difficult to erase.
A second circumstance of biased behavior is evident when street-level bu-
reaucrats respond to general orientations toward clients' worthiness or un-
worthiness that permeate the society and to whose proliferation they regu-
larly contribute. This is one of the most well-grounded generalizations that
can be made concerning client processing. Juvenile court judges determine
sentencing severity on the basis of the apparent worthiness of the defen-
dant. 7 Policemen make decisions concerning citizens on the basis of whether
or not they display respect. Trauma-team personnel tend to work harder to
save the lives of the young than the old, the high-status citizen rather than
the low. 8 Other emergency room personnel make moral evaluations of
clients and treat them accordingly. 9
Where more than one street-level bureaucrat is involved, as in the cases of
courtroom processing or multi-disciplinary assessments of handicapped stu-
dents with special educational needs, it is often the moral worthiness of sub-
jects that is negotiated in these settings. 10
These observations are consistent with the policies of organizations that
tend to focus on morally favored clients instead of those who most require
their services. Agencies for the blind tend to be oriented toward children
and employable adults although most blind people in the United States are
elderly or near retirement. 11 Voluntary hospitals and private social service
agencies tend to relocate or reduce their community services when a new
group populates the neighborhood for which the services were originally
established.
As suggested above, there is every reason to think that the general evalua-
tions of social worth that inform the society will also inform the decisions of
street-level bureaucrats in the absence of strong incentives to the contrary.
Under what circumstances are these general notions of social worth likely to
play a major role? At least three hypotheses seem plausible. First, when
street-level bureaucrats have to choose among clients, and biased selection
will not incur major costs, general notions of moral worth will prove impor-
tant. Studies of emergency room bias suggest that blacks tend to receive
equal treatment when they have severe injuries, and when the emergency
room is not particularly crowded. But when emergency room personnel are
under pressure, blacks with nonserious complaints are discriminated against
as personnel have recourse to diffuse conceptions of worthiness in decision
making.
109
PATTERNS OF PRACTICE
110
Rationing Services: Inequality in Administration
rape." 12 Yet we know that coercion is most effective when it is so complete
that the victim does not resist. Or consider the issues confronting middle-
class judges who must decide whether or not to agree to the petition of a wel-
fare department to remove the children of a low-income parent to foster
homes. When a home. for children becomes <'unfit" under law is hardly a
matter for objective determination.
A third circumstance in which street-level bureaucrats regularly display
biased behavior arises when they are able to act on the view that some
clients are more likely to respond to treatment than others. This category of
favoritism is similar to creaming except that the motivation comes not from
the reward structure of the agency (although this may play a part) but from
the gratification that comes with helping people who are thought likely to
respond to help. Psychologists and psychiatrists often favor verbally ori-
ented, middle-class patients because their modes of therapy are most re-
wardingly practiced with such people. Teachers favor children who assimi-
late information easily because they receive more frequent and positive
feedback. 13
Closely related is the possibility that biased behavior occurs because a
group is regarded as unlikely to respond to intervention. Consider the suit of
the National Association for the Advancement of Colored People against the
New York City Board of Education that charged that black and hispanic
problem children were routinely assigned to special schools, yet white chil-
dren with similar problems were helped in their own schools. 14 Here the
NAACP suggests an instance in which the combined preferences of individ-
ual teachers results in institutional racism.
111
PATTERNS OF PRACTICE
112
Rationing Services: Inequality in Administration
The rule of normality also helps insure that a part of the client population
will be regarded as requiring or able to benefit from intervention, and a part
will be thought of as unresponsive or unworthy of help. In general, street-
level bureaucrats establish expectations of client behavior, both in terms of
performance and in terms of their interaction with the bureaucracy. Devia-
tions from these standards tend to be differentiated. Legal services lawyers
may be more responsive to particularly cooperative clients. 17 Defendants
who fail to show deference to judicial procedures or the agents of the law
may be singled out for particularly harsh treatment. 1 &. An early set of im-
migration regulations designed to exclude people with mental disorders per-
mitted erratic or "hotheaded" behavior if exhibited by Italians for whom this
was regarded as a cultural characteristic (and therefore normal), but
regarded such behavior as grounds for excluding northern Europeans, for
whom this was designated an abnormal characteristic.
It is probably fair to say that clients will always be differentiated in terms
of their perceived relative normality, regardless of how absolutely receptive
to intervention they are. This provides street-level bureaucrats with the in-
surance that they always perceive a set of clients for whom they are neces-
sary. In a school of exceptionally bright children the teachers would learn
quickly who was or was not more gratifying to teach. In a client population
that was not particularly verbal or middle class, psychologists would still be
able to discover quickly who seemed most amenable to treatment. The client
world as perceived by street-level bureaucrats is probably much like the
children judged to be able to profit from tonsilectomies when this operation
was still popular two generations ago. In an experiment recounted by soci-
ologist Eliot Freidson, 174 children out of 38g were selected by a panel of
physicians as being able to benefit from tonsilectomies. However, when
another group of doctors examined the remaining 215 children, 99 more
were judged in need of the operation. And when still another panel exam-
ined the rest (116), nearly one-half were recommended for the procedure. 19
One wonders whether service professionals would ever regard at least a por-
tion of a client population as totally unable to benefit from their intervention.
Like the existence of routines, there is nothing surprising about street-
level bureaucrats' expectations of normal distribution. The critical question
is, again, how are these expectations related to legitimate objectives? The
massive problem of children excluded from school provides a case in point.
Although their expressed purposes are educational, public schools difleren-
tiate extensively on the basis of behavior, excluding large numbers from the
school population through suspensions and other punishments. 20 Similarly
courts differentiate among people charged with the same offense on the basi~
113
PATTERNS OF PRACTICE
not of their street behavior but of their courtroom behavior. In these cases
conceptions of normality operate effectively to deny or restrict client treat-
ment.
5 Self-fulfilling prophecies contribute to the persistence of bias by provid-
ing spurious confirmation of the validity of differentiation. Greater sur-
veillance of adolescent blacks by the police results in their being arrested at a
greater rate than other portions of the population. This tends to confirm that
black young adults are the primary delinquency problem. 21 Sending people
to prison, where they are exposed to experienced criminals,'where they are
labeled as criminal and are later treated as such by society, helps fulfill the
prophecy that they were the kinds of people who required severe punish-
ment in the first place. 22 Hospital staffs through subtle signals induce pa-
tients to display the behaviors of sick people, confirming that they were
mentally disturbed or physically ill in the beginning. 23 The case of schools is
perhaps best known. Teachers who expect achievement from their pupils in-
teract with them in such a way as to bring out their full achievement poten-
tiaJ.24 Those predicted to do poorly, however subtly, are more likely to fail.
It should be no surprise that self-fulfilling prophecies run throughout
street-level bureaucracies. If clients are differentiated they will respond to
that differentiation by accepting in part the implications of the differentiation
for their own identities. They will also respond to the role of client by con-
forming to bureaucracy's expectations concerning client behavior. From this
interactionist perspective it is hardly surprising that differentiation of clien-
teles and the necessity of involvement with bureaucracy lead to the interper-
sonal dynamics we call self-fulfilling prophecies.
6. In the absence of adequate performance measures and in the context of
making significant judgments affecting clients' well-being, street-level bu-
reaucrats depend heavily on subjective assessments of the validity of their
practices. This tendency is strongly supported by the feeling that the work
they do is so specialized that no one else is in a position to criticize or even
comment on their practices. Police, teachers, welfare workers, and other
street-level bureaucrats consider themselves isolated from ordinary citizens
who cannot appreciate the difficulties and abuse that they experience or the
uncertainty of the rewards. 25
Street-level bureaucrats are receptive to information that seems to con-
firm the legitimacy of their differentiation of the client world and thus sup-
ports their patterns of practice. In this they reftect the general psychological
tendency to receive and incorporate information that is supportive of their
world view and to filter out information that appears contradictory. They also
Rationing SeiVices: Inequality in Administration
reflect the general tendency to seek information among peers, who may be
expected to be like-minded.
Street-level bureaucrats are conspicuously prone to scan their environ-
ment for empirical validation of their views. Their conceptions of clients
tend to be consistent with perspectives that exonerate them from responsi-
bility for clients' fate. They are particularly inclined to believe that experi-
ence provides the basis for knowledge in assessing the client world. While
validity by illustration is logically indefensible it is a significant social fact
that influences street-level behavior. We may hypothesize that validity by
illustration ("I know it's true because I once had a client who ... ") will
prevail in proportion to the worker's need to cope with the uncertainties of
decision making and the potential consequences of those decisions. The po-
liceman who draws a gun hasti!y because another officer was recently slain
with a knife when he failed to draw his gun has a powerful argument with
which to defend himself when questioned for abridging department policy.
Undoubtedly there are many street-level bureaucrats who refuse to accept
the perspectives of their jobs that arise in the occupational subculture. Still,
the strength of mechanisms adopted to cope with the work is great precisely
because, ifthey are successful coping devices, they work (by definition). The
need to cope acts as a barrier to anomalous information that might challenge
the routines and orientations that have been developed over time. Changes
in procedures are not necessarily resisted because workers are against
change per se, but because change threatens the existence of coping routines
and orientations that serve to rationalize the work. Similarly, anomalous in-
formation is not heard because it contradicts assumptions that make the job
more rewarding or rationalizes its contradictions.
7 Unsanctioned, persistent differentiation is supported by the racism and
prejudices that permeate the society and are grounded in the structure of in-
equality. Differentiation is intrinsic to street-level bureaucracy, but social
inequality supports it and helps account for the cleavages in terms of which
differentiation takes place. Thus the need to routinize, simplify, and dif-
ferentiate in the context of inequality leads to the institutionalization of the
stereotypical tendencies that permeate the society. Whatever prejudices
street-level bureaucrats as individuals do or do not have, the structure of
their work appears to call for differentiation of the client population, and
thus there is structural receptivity to prejudicial attitudes. The need for
simplification exists, so to speak, prior to the stereotype. The stereotype is
nurtured in a context where it functions to divide up the client population. 26
This does not mean that all street-level bureaucrats are prejudiced or that
115
PATTERNS OF PRACTICE
116
CHAPTER 9
Every social order depends on the general consent of its members. Even the
most coercive of institutions, such as prisons, function only so long as those
affected by the institution cooperate in its activities (even if the cooperation
is secured ultimately by force). Typically, cooperation is neither actively
coerced nor freely given, but, rather, it emerges from the structure of alter-
natives.
In the previous chapters I discussed ways in which patterns of street-level
practice function to ration services. A second general function of street-level
practice is not so much to limit services or choose among clients, but to ob-
tain client cooperation with client-processing procedures. The work that
clients are expected to cooperate with may or may not be consistent with
agencies policy declarations. It will, however, be consistent with street-
level bureaucrats' conceptions of how to process work with minimal risk of
disruption to routine practice.
Street-level bureaucrats' need to control clients as well as the incomplete
nature of that control have been discussed earlier (chapters 5 and 6). Here I
consider selected aspects of practice that commonly contribute to routine
control of clients.
1. Street-level bureaucrats interact with clients in settings that symbolize,
reinforce, and limit their relationship. It is practically a cliche to observe that
the severe appointments of a courtroom, dominated by a bench behind
which a black-robed judge looks down at other courtroom participants, con-
vey the power of the system of laws over the individual. Separate entrances
117
PATTERNS OF PRACTICE
for judges, commands to stand whenever the judge arrives or departs, and
the unintelligibility of the court clerk further contribute to the mysteries of
the courtroom.
Each service setting functions somewhat differently, but in their different
ways each contributes to client compliance. Many offices in which people
seek service are structured to separate clearly the workers from the clients
by means of an imposing information desk. 1 Clients, when interviewed, are
led to "offices" that, lacking partitions, violate privacy by permitting every-
one to view (and listen in on) everyone else's work. Fixed rows of desks in
schools, all facing the teacher, physically represent the demand for order
that teachers and schools require. Like uniforms, settings facilitate the func-
tioning of the bureaucracies by drawing attention to the location of power
and cuing the expectations of clients.
These messages are not accidental. They are fostered by the agencies and
generally consented to by the society. It is interesting to observe in modem
courthouses the extent to which the traditional courtroom setting--dark,
polished wood, the bench, separate entrances, the flags, epigrams celebrat-
ing justice-are retained in otherwise nontraditional architecture.
Consider also the tenacity of setting configurations in other public ser-
vices, and the extent to which departures from tradition appear to be radical.
In the public mind a nontraditional school is simply one without fixed desks.
Are clients important and valued as people? Provide them with comfortable
chairs and sofas on which to sit while they wait, ask them if they are comfort-
able, and reassure them if they must wait that they have not been forgotten.
Are clients of little account? Neglect these considerations and have a small,
cramped waiting room with little attention available. It would be mistaken to
think of service settings as accidental. It is often a matter of policy that public
services are able, or consider themselves unable, to plan for client comfort.
2. Clients are isolated from one another. 2 Public service bureaucracies are
organized so that clients have little knowledge of others in the same position.
Most client processing is shielded from the scrutiny of other clients. Isolated
clients are more likely to think of themselves as responsible for their situa-
tions. They are unlikely to see their condition as a reflection of social struc-
ture and their treatment as unacceptable.
When client processing is done in public, the impression is accurately
conveyed that clients are competing with one another for the attention or
favor of street-level bureaucrats. As suggested earlier, in the brutal realities
of triage, clients perceive that they gain special treatment or the attention of
workers only at the expense of other clients. The bureaucratic defense
against special treatment is also germane here: "If I give it to you I would
118
Controlling Clients and the Work Situation
have to give it to everyone." In street-level settings in which clients do know
each other-in schools, mental hospitals, prisons-<!lient control is fostered
by the competitive systems of rewards, fostering among clients individual
orientations rather than collective solutions to problem solving.
Street-level bureaucracies tend to resist organization by clients when it
occurs. They tend to regard client organizations as unnecessary, frivolous,
likely to be irresponsible, or not representative of clients' true interests.
There are no objective measures of the validity of such assertions. From
some perspectives any or all might be true. However, these assertions are
most usefully regarded as defenses against client organization, intended to
diminish their influence among potential recruits or third parties whose sup-
port is sought, or to lay the groundwork for an intransigent official response.
In the past decade prison inmates, black high-school students, and welfare
recipients all have been regularly subject to such official responses when
they have attempted to organize.
Public officials often prefer to suppress or disorient client organizations
because they can never be sure at what point they will peak or major conces-
sions will be required. However, one lesson learned well by public officials
during the past ten years is that it is often possible and desirable to encour-
age client organizations in order to provide a buffer between individual
clients and the agency. Lacking substantive powers or the resources to act
effectively, client organizations often provide the appearance of access while
actually influencing only those areas in which policy decisions do not materi-
ally affect agency behavior.
3 The services and procedures of street-level bureaucrats are presented
as benign. 3 Actions affecting clients are always taken in their best interest.
Clients are expected to be grateful for benefits they receive. Where street-
level bureaucracies constrain clients who are not regarded as guilty-as in
schools, hospitals, and noncriminal arrests by the police (e.g., apprehension
of alcoholics}-the ideology of benign intervention is particularly necessary
to justify practices of questionable value to both client and worker groups.
When combined with clients' deference to the more extensive education,
training, and expertise of street-level bureaucrats, the ideology that street-
level bureaucrats' intervention is in the interest of clients appears to be a
particularly important instrument of control. 4
4 Clients must come in for service. With a few important exceptions
street-level bureaucracies require clients to appear for service, rather than
have workers go to clients. In part this is a matter of efficiency. As suggested
earlier, the monetary costs of providing service would increase dramatically
if the costs of having to wait were borne by workers rather than clients. How-
119
PATTERNS OF PRACTICE
121
PATTERNS OF PRACTICE
to perceive that clients are significantly different from each other, except in
extreme cases. Only clients who are particularly uncooperative, and those
who are particularly able to assist in managing their own cases, tend to re-
ceive differential responses. Welfare workers also fail to differentiate among
clients. Like legal services lawyers, their interaction with clients is so highly
structured that variations in treatment are not apparent. 7
This is not to say that legal services clients or welfare recipients always
present unique situations. On the whole there is probably a good deal of sim-
ilarity in the situations street-level bureaucrats confront. However, the rou-
tinization of inquiry minimizes the extent to which street-level bureaucrats
can discover unique circumstances requiring flexible responses. Thus we
have the ingredients for another self-fulfilling prophecy. In the expectation
that most clients will fall into previously defined categories, bureaucracies
follow search procedures based on that expectation. Having constricted the
kinds of information they receive, street-level bureaucracies find confirma-
tion that, indeed, clients tend to fall into certain well-defined categories.
6. When control of clients is problematic yet critical to task performance
or personal safety, interactions between citizens and street-level bureaucrats
are dominated by control routines. Some street-level bureaucrats cannot
depend on the setting, or on a set interview process, to control clients.
These workers will develop routines to make client control a precondition of
the interaction.
In teachers' interactions with children and police interactions with sus-
pected offenders, symbols of authority and prior socialization help to insure
client compliance with street-level bureaucracy. However, these social con-
trol mechanisms are often insufficient. Teachers and police officers both
must act immediately to secure the cooperation of the people with whom
they are engaged. Inner-city school teachers, for example, consider main-
taining discipline one of their primary problems. It is a particularly critical
issue in providing educational services when "keeping them in line" and
avoiding phyo;ical confrontations consume a major portion of teachers' time. 8
Even under less threatening circumstances elementary school teachers are
urged to "routinize as much as possible" 9 in order to succeed.
Police rookies are informally socialized by veterans to be tough as a pri-
mary requisite of their training 10 and are regarded as dependable partners in
large measure by their willingness to use force when necessary. Police
officers have most call for mechanisms that will secure compliance with their
authority. They depend on such routines for protection from what might
evolve into a life-and-death conflict without resorting to a show of arms or
other unacceptable deterrents. One way in which they routinize their ap-
122
Controlling Clients and the Work Situation
proach to interactions is to maximize their safety in the event of an attack.
Thus, for example, police officers physically position themselves to be in the
most advantageous situation if an attack materializes, even if the probability
of an assault is low. 11
Another primary orientation of police routines is to manipulate their pub-
lic image as ready to use violence if necessary. As one observer describes it:
In most threatening situations, the officer attempts to maintain his edge by managing
his appearance such that others will believe he is ready, if not anxious, for action. The
policeman's famous swagger, the loud barking tone of his voice, the unsnapped hol-
ster or the hand clasped to his nightstick are all attitudes assumed to convey this im-
pression. Decisiveness is readily apparent in such a posture, although the officer
himself may have little, if any, idea of what he is about to do. 12
123
PATTERNS OF PRACTICE
124
Controlling Clients and the Work Situation
ever, they cause great consternation among welfare workers who fear that
their actions might create a precedent for evading the control procedures of
the department. Like parents who can allow exceptions to family rules only
with a warning to their children that negates any implication of generosity or
flexibility, welfare workers typically make such exceptions only with a warn-
ing to their clients that they would be well-advised to learn to read and
write. 17
An analogous process of chastising clients toward whom they are other-
wise being generous is observed among judges who deal out probationary
sentences instead of incarceration. Judges sentencing offenders to probation
tend to chastise and moralize more than when they sentence offenders to
jail. 18 It is as if the judge, guarding against the possibility that the decision
will be called into question by the later behavior of the offender, must com-
pensate for leniency by applying a verbal equivalent of punishment.
Husbanding Resources
125
PATTERNS OF PRACTICE
rooms. It is also true of welfare offices, schools, and other agencies in which
the work pace would seem more predictable. A client may appear who
presents a particularly complicated processing problem and a few children
may prove particularly disruptive on a given day. When combined with
unexpected but regular breakdowns in agency functioning-such as the
need to cover for a co-worker, or the intrusion of some new regulations
requiring implementation-these unpredictable circumstances require that
street-level bureaucrats have some reserve capacity to respond.
Workers faced with unpredictable surges in work load will attempt to
secure safe time that they can deploy if necessary, time that usually functions
to cushion the work day. "Where workloads or resource supplies fluctuate,
the individual is tempted to stockpile .... " 19 This is the significance of the
police dispatcher's tacit approval of officers who, upon completing an assign-
ment, delay calling in until they have completed leftover paperwork.
Cushioning the work day is rationalized in public services in which there is a
well-recognized need for reserve capacity. Police departments and
emergency rooms may be expected to deploy resources for peak load duty,
but only within certain reasonable limits. In these services there must
always be a planned excess capacity if they are to function effectively. In con-
trast, public services without generally recognized emergency functions are,
if anything, expected to operate above capacity. When schools operate on
double shifts or welfare workers have excessive case loads these institutions
are not considered to be malfunctioning except by a small group of
concerned and attentive citizens.
In such situations street-level bureaucrats have a strong incentive, when
they can, to let the work expand to fill the amount of time available. This
Peter Principle is commonly perceived to result from laziness, inertia, and
inability to plan. While it may be related to these things, it is also rooted in
the bureaucratic need to cope with the unpredictability of demand.
Workers recognize that any sign that they are not working to full capacity
will be greeted with additional assignments, as there are always additional
assignments to be made. Thus there is no incentive to complete an inspec-
tion or an interview quickly, because the time liberated by efficiency will be
reassigned to conduct another inspection or interview. This is not the case
when workers have quotas to meet or business appointments to conduct. In
these cases workers are rewarded by processing assignments more quickly
since they will be able to enjoy or utilize for other purposes the time liber-
ated by their efficiency.
The validity of conceiving street-level bureaucrats as acting to conserve
work time may be judged in part by the vehemence with which workers seek
Controlling Clients and the Work Situation
to protect their current work loads. Case loads are invariably too heavy. Re-
sponsibilities are always burdensome. These arguments are difficult to in-
spect rationally. Efficiency specialists will seek to disclose an untapped
horde of resources and appropriate them for the agency. Workers' represen-
tatives will demonstrate the good uses street-level bureaucrats make of their
slack time, if they have any-doing paperwork, preparing for assignments,
or taking a breather so as to remain effective-and will demand compensa-
tion if this work must be done while workers are off-duty. For clients slack
time remains a target of apparently greater available resources, of which
both individual clients and groups of clients might take advantage.
The possibilities that workers are overworked and slothful remain. But it
is more helpful to see the problem as structural. Workers have no protection
against the accretion of responsibilities because of incessant demands and
the irrelevance of performance measures of service quality. Given the need
for reserve capacity to meet unpredictable demands, the husbanding of
resources follows.
At the organizational level there are some conspicuously typical ways in
which problems of the need for reserve time are handled. Organizations
relieve their staffs of the need to treat irregular processing problems by de-
veloping special units and special procedures to process problem cases.
These mechanisms are discussed at some length below (pp. 133-139).
Another way in which street-level bureaucrats conserve resources is to
create the conditions for making decisions as free as possible from the im-
plicit pressures of affected parties. For example, when education personnel
in Massachusetts were required to design an educational plan for all children
with special educational needs in the presence and with the active partici-
pation of the parents, these meetings were often preceded by private meet-
ings of the evaluating teams. It is reasonable to infer that the teams met
without the parents in order to present the parents with a united front and to
avoid having to make a decision on the spur of the moment in the presence of
outsiders. 20
Another illustration is provided by Boston housing inspectors. These
workers are required to give complainants a copy of the violations they
record following an inspection. This regulation was introduced to make the
actions of the inspectors visible to tenants who complained and to leave a
record of the visit. However, inspectors are not required to leave with ten-
ants a copy of their "inspector's report," which sets out the actions taken by
the inspector after leaving the premises, including the actions taken on the
violations found. Thus the tenant may receive the impression that the in-
spector's visit was fruitful. In reality the inspector may decide not to take
127
PATTERNS OF PRACTICE
legal action on the violations, and this decision is not taken when the tenant
can review or have access to the process. Moreover, if the inspector decides
that there is no cause for action on the violations, the tenant would only learn
about it if she or he inquired later why no action was forthcoming. 21
Similarly, lower-court judges can be observed to reserve decisions for
later announcement when they think one of the parties to a case would be
highly reactive, or if a worthy but guilty party to a case would be disap-
pointed by a decision.
The desire to make decisions about people in private, particularly when
they are likely to be negative or disappointing, is entirely understandable.
Yet public policy often calls for street-level bureaucrats to make decisions in
public precisely so that they can be exposed to, if not influenced by, the
presence of those affected by the decisions. Such policy incorporates the
theory that clients are likely to become more a part of bureaucrats' reference
groups if they are present at times when decisions are made. Thus it is a mat-
ter of concern that street-level bureaucrats often are able to shift the field of
decision making to a place where clients cannot intrude. The associated ten-
dency to obscure the locus of decision making so that no individual need be
confronted by disappointed or aggrieved clients similarly threatens client
interests.
A final set of practices that function to husband resources have in common
the effective transferring of decision-making responsibility about clients to
other public workers. Street-level bureaucrats do this by permitting lower-
level functionaries to exercise discretion in their place (screening), by ac-
cepting the judgment of others so that independent assessments need not be
made ("rubber-stamping"), and by referrals.
SCREENING
Most people-processing organizations have a formal role for workers who
stand as buffers between street-level bureaucrats and clients. Although
street-level bureaucrats formally make the critical decisions about clients'
status or provide the services, there remains an important formal role for
other workers. ;I'his is to provide information to clients, to determine the
proper slot for clients when discretionary decisions are minimal, or to pro-
tect street-level bureaucrats from inappropriate client pressures (as defined
by the agency).
Thus street-level bureaucracies commonly employ receptionists, clerks,
secretaries, and other facilitators to provide information in person or over
the telephone, to assist clients in locating the proper place in the bureau-
cracy, and to make informal judgments as to whether they should persist in
128
Controlling Clients and the Work Situation
seeking help. This statement is equally applicable to the receptionist in a
legal services office, a health aid in a VA hospital, an operator in a 911 police
emergency call system, and a secretary of the local public housing authority.
The role of screener would not be cause for comment if screeners per-
formed their jobs as they are defined in theory-that is, making decisions in-
volving minimal discretion. However, in important respects screeners often
come to function as street-level bureaucrats, exercising discretion in impor-
tant areas of people's lives, although without the authority to do so.
Emergency room registration clerks determine the order in which patients
are seen by doctors, and even whether patients will be seen at all. 22 Secre-
taries can be helpful or unresponsive to potential clients, by their manner
conveying whether or not the agency is likely to be receptive, and by help-
ing or not helping potential clients characterize their situations in ways that
will gain them favor. Like registration clerks in emergency rooms, the way
secretaries characterize cases affects the ultimate processing by street-level
bureaucrats. 23
Perhaps an extreme example is provided by the public-housing applica-
tions clerk who identified prospective tenants she considered desirable,
placed them in projects before vacancies became generally available, and in-
formally rejected acceptable applications by placing them in a special file in
the back of the applications file drawer. She did all this while working for an
agency whose formal procedures called for fair and nondiscretionary behav-
ior on her part. 24
Wherever workers encounter the public they are in positions to play the
gatekeeping functions of determining eligibility, conveying information, and
presenting the face of the agency to clients as benign, indifferent, or hostile.
Thus whether the low-level worker's tasks are tightly circumscribed, as in
VA hospitals or police emergency telephone systems, or they present con-
siderable opportunity for the exercise of informal discretion, as in public
housing, these agency buffers can vitally influence citizen access to public
benefits.
RUBBER STAMPING
A different kind of screening is used as the basis for decision making by
street-level bureaucrats who routinely adopt the judgments of others as their
own. Sometimes the views of others determine the actions street-level work-
ers will take; sometimes they will determine the effort workers will devote.
Popularly we call this "rubber stamping," although the process is often more
complicated than this pejorative label implies.
For example, judges commonly accept the decisions of police officers or
129
PATTERNS OF PRACTICE
probation officers in lower-court criminal cases and more or less ratify these
decisions in their determinations. 25 In some domestic relations courts the
recommendations of social workers on the placement of children in con-
tested divorces almost invariably provide the guidelines for judicial action,
although their reports are supposed to be only advisory. The same has been
observed of adoption agency workers and judicial consent to placement. 26
According to Jerome Carlin, in many specialized judicial settings that deal
with the poor "it has become common practice to delegl}te authority for
decision making to administrative personnel: referrees, commissioners, pro-
bation officers, medical examiners, marriage counselors, and others." 27
A particularly significant delegation of authority has taken place in com-
mitments to mental hospitals and in other situations in which mental com-
petence is at issue. Commitments or findings of incompetence often may
take place without hearings or, for that matter, any other proceedings in
which contradictory evidence might conceivably be presented, simply on
the basis of petitions for hospitalization or the reports of physicians. 28 These
practices, however modernized or reformed to some degree in recent years,
provides insight into the abrogation of discretionary decision-making func-
tions under some circumstances.
Judges are not the only ones who accept the recommendations of others in
making their authoritative decisions. Teachers use the judgments of chil-
drens' previous teachers in their informal classroom track assignments, as
early as the second grade, according to one study. 29 Public defenders make
judgments about the worthiness of offenders on the basis of the charges
made by police officers. 30 Emergency room doctors accept a clerk's judg-
ment as to whether an individual requires immediate assistance or is a drunk
and therefore should be treated after other patients, although the "drunk"
may not smell of alcohol and may have serious medical requirements that
would otherwise command the doctors' attention. 31 Perhaps the most me-
chanical application of previously applied labels is that of emergency room
personnel who give immediate attention to all patients who arrive in an of-
ficial vehicle, such as a police car, regardless of their condition. 32
It is fairly easy to understand why bureaucrats would consistently accept
the judgment of others in making determinations. Street-level bureaucrats
confront problems in which they must make significant decisions about peo-
ple and complex situations without being able to interrogate people fully or
investigate the background of their claims. The assertions of other profes-
sionals, who are assumed to know their jobs and are charged with responsi-
bility for making appropriate assessment in their own work, provide signifi-
130
Controlling Clients and the Work Situation
cant and legitimate cues to decision making in the absence of other sources
of information.
It is entirely rational to depend upon the cues of respected others in mak-
ing decisions. People do this all the time-when asking friends to recom-
mend a mechanic, or a movie, or a mover. Unfortunately, what is rational
private decision making may subvert public policy. Judges, rather than po-
licemen, probation officers, or social workers, are charged with judicial deter-
minations because they are theoretically in a better position to seek and hear
information from all sides, and procedurally in a better position to protect
the participants in a case. From a limited perspective this may be the best
they can do. But when judges pass on responsibility they negate the theoret-
ical safeguards their responsibilities represent in favor of the safety of expert
or informal opinion.
Public policy is also subverted because street-level bureaucrats are gener-
ally obliged to make decisions based upon the case at hand. When clients are
presented with or come accompanied by labels that predict the treatment
they will receive, they do not obtain a response to the case at hand but rather
to the stereotype their label evokes. Thus the "troublemaker" in school, the
"drunk" in the emergency room, and the "rotten apple" in juvenile court re-
ceive responses to their labels and not to the behavior or circumstances that
brought them into association with public agencies in the first place.
Still another difficulty with processing clients through prior identification
by others is that the professionals whose decisions are effectively substituted
for the decisions of those formally charged with authority in these cases are
themselves subject to the decision pressures of street-level bureaucrats. The
probation officers or social workers with many investigations to make and
little chance of making them thoroughly take shortcuts of characterization
and judgment that are similar but not identical to those made by the officials
to whom they report. Thus the social worker is horrified to discover that her
ambivalent, highly tentative report on placing children in a divorce proceed-
ing, rendered under great pressure, highly qualified to reflect the uncer-
tainty of findings, is taken as authoritative and as an appropriate basis for ac-
tion, despite the pleas in her report for further, intensive probing. Hers is
the only information that the judge has to make a decision. The feelings and
actions of both social worker and judge are understandable. But however
understandable, the interests of family members in a fair and humane deci-
sion may be jeopardized. 33
131
PATTERNS OF PRACTICE
REFERRALS
A final set of practices operating to conserve resources is associated with
referrals. Referring a client from one agency to another obviously serves the
client's interest when there is an identified, specific client need and re-
sources are available from the receiving organization. However, there is a
class of referrals which, whatever its contribution to client well-being, ap-
pears to function more to process heavy case loads in resource-poor agencies
than to fulfill specific client needs. Street-level bureaucrats may make refer-
rals as one of the least costly ways to process clients without providing ser-
vices. Thus agencies may maintain benign images of helpfulness and service,
without explicitly having to tum clients away.
This use of referrals is partly a result of the extraordinary demand for
resources relative to the supply. Public agencies, responsibly seeking to
meet clients' needs, attempt to link them with other agencies when their
own resources become swamped. This works to the satisfaction of all when
resources are available in other agencies, but it turns into a referral merry-
go-round when other agencies become similarly inundated. When the Rox-
bury (Boston) Multi-Service Center opened its doors, it expected to provide
black community residents with links to other social agencies to which they
might not otherwise go. However, the agency shortly experienced as many
referrals from other agencies as it was making to other agencies. It may be
inferred that other Boston social agencies, under-staffed relative to the de-
mands on them, saw the new Multi-Service Center as a resource that they
might now exploit. 3 4
Referrals also may represent a way in which agencies protect themselves
by providing symbolic service when actual services are not available. Jeffry
Galper puts it audaciously when he suggests the analogy of the parking
problem in large cities: "there are never enough spots for all in need, but at
any one time some cars are cruising the streets, looking. Referral is a way of
dealing with clients in need without really dealing with them. 35
Referrals also have some of the qualities of court delays and waiting lists.
More people can be accommodated into the service structure at orie time, al-
though no more service is actually provided. And referrals can result in in-
ducing people to stop seeking services because they consider their need less
important now relative to the costs, or they have been encouraged to resolve
their problems on their own. Whatever explains the drop-off from the refer-
ral net, it functions to some extent to ration the community services
available.
132
Controlling Clients and the Work Situation
1 33
PATTERNS OF PRACTICE
decision may evoke client hostility, the worker can process the case through
a pressure referral. Thus the worker's legitimacy is partially protected by the
availability of a channel that places responsibility for difficult decisions in the
hands of others.
Workers can use the availability of pressure specialists to enhance the
prospects of favored clients. For example, welfare workers often take plea-
sure in artfully presenting cases to supervisors in such a way that they are
likely to endorse the worker's judgment. Or street-level bureaucrats can
scuttle clients' prospects without clients' knowledge by giving the appear-
ance of bureaucratic neutrality but privately providing damaging information
to supervisors.
The possibility that decisions can be appealed also enhances the legiti-
macy of the bureaucracy to the client. For this to work on a sustained basis,
however, two conditions must be met. First, and quite obviously, it must
look like channels for appeal are open. Second, and less obviously, these
channels must be costly to use, rarely successful, and, if successful, certainly
not well publicized. The reason for this is simply that if appeals channels
were inexpensive to use or likely to be successful they soon would be used by
clients seeking increased benefits or a favorable disposition. The channels of
appeal would soon be clogged, and the manifest unfairness that some clients
receive more than others because they sought more would undermine the
system. 37
Thus appeals ordinarily require long delays, the services of advocates,
complicated administrative procedures associated with filing, and general
hostility from the challenged agency. 38 Recent innovations responsive to
client pressure often require public agencies to publish the requirements for
appealing and inform clients of their rights to appeal, provide responses
within a specified time period, and offer counsel to clients seeking appeals.
These innovations still require considerable determination and energy from
individual clients.
Public agencies also seek to insure that appeals cannot be sought collec-
tively. The appeals process can function so long as a single client cannot gain
redress for a class of clients. So long as individual clients cannot win benefits
for groups, public agencies can ration the claims oflarge numbers of clients
in many ways, and thus gain protection from an inundation of client
demands.
These observations are generally supported by examining the volume of
appeals in public agencies. For example, through the early 1g6os there was
almost a total lack of appeals from welfare decisions, although federal law
required each state to establish an appeals procedure. In New York City,
134
Controlling Clients and the Work Situation
where a relatively liberal welfare environment prevailed compared to the
rest of the country, only 15 appeals were taken in 1964, although half a
million people were on welfare at the time. 39
Appeals can also be discouraged by the high probability that they will not
succeed. Allegations of police brutality are rarely made through official chan-
nels because of the conviction that they will not receive a sympathetic hear-
ing from the officers who sit on the hearing boards. In Rochester, for ex-
ample, where 102 complaints alleging "unnecessary force" were registered
in the five- to seven-year period after 1965, only two were upheld by the
police internal inspection office; of the 368 alleging unnecessary force and
other improper behavior, forty-six were sustained. 40
At times street-level bureaucracies institutionalize the pressure specialist,
creating a special unit to deal with cases with which the agency is generally
troubled. In many municipalities landlords are not vigorously prosecuted for
housing violations, in part because judges do not give high priority to hous-
ing cases when cases of apparently greater public urgency-assaults and nar-
cotics, for example-are also brought before them. 41 However, at times the
failure of the courts to prosecute landlords vigorously threatens to under-
mine the legitimacy of the court systems. In such cases local housing courts,
like the one established in Boston, can segregate these troublesome matters
while reserving court time for the regular case load.
Tactical patrol forces provide police departments with special capacity to
allocate officers to high crime neighborhoods and situations with high poten-
tial for violence. Special classes for disruptive students and those with learn-
ing disabilities absolve teachers from a need to deal with these control prob-
lems. This has significant consequences for children assigned to these
classes, as well as for mainstream students deprived of their presence. Even
if a justification for such segregation can be mustered, the highly subjective
nature of classifying students for such classes draws attention to the organiza-
tional rather than the educational functions they serve.
A typical response of many public agencies to the claims generated by mi-
nority and women's rights movements has been to establish special units to
hear citizen complaints and to take responsibility for institutional change in
these areas. Police departments have established internal review boards
(sometimes with outside citizen participation) and community relations units
to present a sympathetic face to the black community. Public school systems
have hired community relations specialists and affirmative action officers to
take responsibility for the complaints of minorities and women and to articu-
late agency perspectives consistent with the interests of these groups. These
steps have contributed to increased minority and female employment in the
1 35
PATTERNS OF PRACTICE
EMERGENCIES
The most common structural device for managing the consequences of
routine is the emergency. Emergency practices, whatever their other func-
tions, solve major service dilemmas for street-level bureaucracies.
Most people are generally familiar with some emergency procedures.
Hospitals assign emergency status to people with certain medical conditions
and accordingly take extraordinary actions on their behalf. Ancillary health
protection agencies such as ambulance services and fire departments take
extraordinary actions when an emergency is identified. Police dispatch as-
Controlling Clients and the Work Situation
signs degrees of priority to calls in which intervention would most likely save
lives or result in the apprehension of felons.
Emergencies are not confined to the reactive, life-saving public services
Americans have become so familiar with through television. New welfare re-
cipients can receive emergency grants upon applying; continuing recipients
can receive them if they are burned out or experience other disasters. Hous-
ing inspections may be expedited if they are classified as emergencies, and in
some cities housing may even be repaired by public agencies if an
emergency is judged to exist. Public housing applicants receive priority if
their need for housing has emergency status. Mental health clinics are in-
structed to take only clients with acute problems who present a danger to
themselves or others. Legal services offices sometimes accept only clients
who have emergency legal needs. It is the rare aid-giving agency that does
not provide emergency assistance, the rare service-providing agency that
does not admit emergency cases to its rolls.
Yet what is an emergency? There would probably be some consensus that
the word denotes a situation that requires prompt attention, threatens the
sustained existence of the subject, and calls for extraordinary actions on the
part of others. 44
However, the use of emergency treatment categories or assignment of
clients to them may not reflect these circumstances. Police emergencies
calling for preemptive responses include situations in which an emergency is
designated primarily because it seems likely that offenders could be caught.
Yet the potential for apprehending criminals is not necessarily related to life
saving. The highest priority is informally assigned to situations in which
officers are the ones whose lives are threatened. Here the highest priority
depends on the occupation of the subjects.
It is relatively easy to find other examples in which emergencies depart
from this definition. As indicated previously, emergency rooms accord spe-
cial treatment to patients who arrive in police vehicles, regardless of their
medical condition. Public housing officers and welfare workers can show fa-
voritism by instructing selected clients how to be categorized as emergen-
cies. New York City developed an emergency response to deteriorated hous-
ing conditions that accorded priority to conditions brought to the attention of
city agencies by rent strikers and other tenant groups. 45 Here emergency
status was affected by who sounded the alarm.
Clearly the category of emergency in public services is organizationally
and situationally determined. A condition is emergent in public services
only if it is called so by an authoritative agent. Patterns usually emerge to
give shape to emergency practice, so that by convention one can often tell
1 37
PATTERNS OF PRACTICE
139
CHAPTER 10
The Client-Processing
Mentality
The drill sergeant who insists that soldiers stand tall, keep their eyes
straight, and march in precision achieves results without knowing the state
of mind, predispositions, or previous military experience of the recruits. He
is untroubled by the needs of individuals and is at ease with mass processing.
Street-level bureaucrats are not so favored. Their work involves the built-in
contradiction that, while expected to exercise discretion in response to indi-
viduals and individual cases, in practice they must process people in terms of
routines, stereotypes, and other mechanisms that facilitate work tasks.
Workers defend these patterns psychologically. They regard their adapta-
tions to the job not only as mechanisms to cope with resource limitations,
but also as functional requirements of doing the job in the first place. Thus
what to critics seem to be compromise solutions to resource constraints may,
from the workers' perspectives, be desirable and necessary components of
the work environment. To attack the routine is to appear to attack the struc-
ture. Clients who challenge bureaucratic routines are taught this lesson
when administrators act to control them or respond defensively to questions
about agency procedures.
However, this does not entirely explain how workers cope or exhaust the
types of psychological adaptations apparently required by these jobs. For
one thing, it does not explain how street-level bureaucrats rationalize the
discrepancy between service ideals and service provision. At least two addi-
tional perspectives on the psychology of street-level work must be consid-
ered in accounting for street-level bureaucrats' persistence and relative job
satisfaction.
The Client-Processing Mentality
First, street-level bureaucrats modify their objectives to match better
their ability to perform. Second, they mentally discount their clientele so as
to reduce the tension resulting from their inability to deal with citizens ac-
cording to ideal service models. In short, street-level bureaucrats develop
conceptions of their jobs, and of clients, that reduce the strain between
capabilities and goals, thereby making their jobs psychologically easier to
manage. 1
This is particularly significant because street-level bureaucrats' views of
their work, and of clients, are matters of great public concern. Street-level
bureaucrats are often accused of being biased against particular racial or eth-
nic groups or they are thought to be particularly cynical or unreliable in ful-
filling obligations toward particular social groups. The proposal that workers'
attitudes in large part are formed in response to their work setting contra-
dicts some popular views. Popular wisdom often identifies the source of
workers' attitudes toward clients and their jobs in prejudices acquired in up-
bringing and social background. Such perspectives lead to recommendations
to hire better educated personnel or provide further education and training
in public and human relations.
All too often such perspectives fail to take account of the influence of
street-level bureaucrats' work on their attitudes. It is apparent that street-
level bureaucrats change their attitudes from the time they are recruited to
the time when they begin to experience work problems. Differences in the
class backgrounds of recruits tend to disappear in training and trainee sociali-
zation. 2 Furthermore, there is evidence that educational background, which
is closely related to class, is not an important predictor of the attitudes of
workers who experience extreme job stresses. In this connection, sociologist
Eliot Freidson has reviewed studies relating doctors' educational back-
ground to performance and concludes: "There is some very persuasive evi-
dence that 'socialization' does not explain some important elements of pro-
fessional performance half so well as does the organization of the immediate
work environment. 3
This is not to say that biases toward clients do not intrude in street-level
work. However, focusing on the social backgrounds or experiences of work-
ers will not yield a persuasive theory of bias in street-level bureaucracy.
Such a theory should account for the development and persistence of atti-
tudes as well as their direction.
Taking a different view, the origins of bias in street-level bureaucracies
may be sought in the structure of work that requires coping responses to job
stress. Attitudinal developments that redefine the nature of the job, or the
nature of the clientele to be served, function in this way. Considering the
PATTER~S OF PRACTICE
structure of work helps explain the persistence of biases and the difficulties
inherent in interrupting them.
However, the content of coping responses may well reflect the prevailing
biases of the society. The need for biases may be rooted in the work struc-
ture, but the expression of this need may take different forms. Stereotyping
thus may be thought of as a form of simplification. While simplifications are
mental shortcuts (of many different kinds) that summarize and come to stand
for more complex phenomena, stereotypes are simplifications in whose va-
lidity people strongly believe, and yet they are prejudicial and inaccurate
as summary characteristics for groups of people with nominally similar
attributes.
This approach to analyzing the client-processing mentality detaches the
existence of attitudes toward clients and jobs from the content of those atti-
tudes. It suggests that attitudinal dispositions will be rigid or flexible in large
measure according to the degree they help workers cope with job stresses.
On the other hand, it suggests that workers' attitudes and resulting behavior
may be challenged and helped to change if: incentives and sanctions within
the structure of the job encourage change; the structure of the job is altered
to reduce workers' needs for psychological coping mechanisms; it can be
shown that workers can cope successfully with job stresses without depend-
ing upon undesirable simplifications; efforts are made to make simplifica-
tions conform to actual job requirements rather than to unrelated biases.
These general guidelines are grounded in recognition that the persistence of
inappropriate attitudes is related to the work experience, and they can best
be helped to change by focusing attention on the requirements of work.
The following sections treat in greater detail the tendency of street-level
bureaucrats to cope with job stresses by modifying their conceptions of work
and their conceptions of the clientele to be served. At the same time they
show the relationship between attitudinal coping responses and the patterns
of practice that the attitudes support.
achieved. The cynical view is that public workers have very little incentive
to perform. However, while some street-level bureaucrats may retire on the
job, the vast majority continue to be reasonably dedicated to occupational
objectives as they come to define them. 6
In addition to the usual material and psychological incentives operating on
the job, street-level bureaucrats often enter public service with some inter-
est in client-oriented work, embrace professional orientations that call for al-
truistic behavior toward clients, and continually interact with clients, thus
regularly confronting client characteristics and concerns. Mqreover, street-
level bureaucrats do not abandon agency objectives entirely because the
discretionary nature of their jobs and the organizational milieu in which they
work encourage them to develop private conceptions of the agency's objec-
tives. They strive to realize these moditied objectives and measure their day-
to-day achievements in terms of them. They rationalize ambiguities and con-
tradictions in objectives by developing their own conceptions of the public
service (which they may share with other workers). Taking limitations in the
work as a fixed reality rather than a problem with which to grapple, street-
level bureaucrats forge a way to obtain job satisfaction and consistency be-
tween aspirations and perceived capability.
Accepting limitations as fixed rather than as problematic is significant for
two reasons. First, it discourages innovation and encourages mediocrity. It is
one thing to say that resources are limited, another to say that the practices
arising from trying to cope with limited resources are optimal. Yet the ten-
dency to equate what exists with what is best is strong when patterns of prac-
tice must be defended psychologically to avoid confrontations with work fail-
ures.
Second, as I have argued, organizational patterns of practice in street-
level bureaucracies are the policies of the organization. Thus, workers' pri-
vate redefinition of agency ends result directly in accepting the means as
ends. Means may become ends in other organizations, but lower-level work-
ers rarely have as much influence on the drift in goals as in street-level bu-
reaucracies.
144
The Client-Processing Mentality
better fit between their capabilities and goals, 7 so too workers can and do
modify their conceptions of the job in order to close the psychological gap
between capabilities and objectives. Thus judges may be oriented toward
punishment and deterrence or corrections and rehabilitation. Teachers may
be oriented toward classroom control or toward cognitive and personality
development. Police officers drift toward concerns with order maintenance
or law enforcement. 8 Possessing a simpler concept of the job than the one
theoretically prevailing in reality, street-level bureaucrats are able to fashion
an apparently more consistent approach to their work.
Street-level bureaucrats also impose personal conceptions of their jobs
when they make superior efforts for some clients, conceding that they cannot
extend themselves for all. At times this perspective results in favoritism
toward certain social groups, but it may also apply without group bias. A case
in point is the public defender who must select only a few cases to push to
trial, settling the others as best he or she can. 9 Teachers similarly rationalize
their inability to pay close attention to all children by drawing special satis-
faction from the progress of children who do receive particular notice.
In these cases efficiency is still the norm and effective triage is again the
ideal. But the benefits gained from modifying goals to make them consistent
with serving a few, when not all can be served well, are not public benefits.
On the contrary they are enjoyed mostly by the workers (and presumably by
the clients who receive special attention). Moreover, they are not open to
popular judgment or normally available for policy analysis. The individual
street-level bureaucrat is not, in a sense, free to abandon private conceptions
of the job without taking on still more of the tensions that go with it. Because
these personal conceptions are adaptive responses they tend to be held
rigidly and are not open for discussion.
The patterns of practice developed by individual workers often only make
sense in the private conception of the job held by the worker, while super-
visors and the public still expect allegiance to a more complex set of goals.
For example, a police officer who fails to make an arrest upon observing an
unlawful incident may strike an observer as negligent. But if the officer pri-
vately understands his or her job to be one of maintaining order and commu-
nity harmony, with law enforcement in the neighborhood a secondary mat-
ter, this behavior may be acceptable according to the officer's private
definition.
In the same way, a teacher who spends a great deal of time with a few
students will not consider fair any criticism of this practice if he or she
defines the job as, at best, the provision of sufficient attention to a select
145
PATTERNS OF PRACTICE
group. It is difficult to investigate conceptions of the job and trace their rela-
tionship to performance. Yet this may be necessary if one would try to
reorient street-level bureaucrats in their work.
Private conceptions of the job have their counterparts in official policy. In
some cases agencies themselves solve workers' problems by imposing a par-
ticular orientation on the work. At other times, the adaptive defensive atti-
tudes of street-level bureaucrats toward their jobs are incorporated in the
service orientation of their agencies although still officially unsanctioned.
Thus the staff of some schools develop collective perspectives on their work
and some police departments develop a shared view of patrol practices, con-
trary to the preferences of supervisors. Recruitment of like-minded people
to the service contributes to collective adaptation to bureaucratic stresses by
excluding staff members who would challenge work-force goal consensus. 10
SPECIALIZATION
Specialization of function in bureaucracy is usually treated as fostering ef-
ficiency, permitting workers to develop skills and expertise and concentrate
attention on their work. For some analysts specialization is synonymous with
modem bureaucracy . 11 Specialization is frequently and increasingly char-
acteristic of street-level bureaucrac.~s. Welfare departments separate social
services from eligibility determinations. Legal services agencies separate in-
dividual client servicing from law reform units. Schools breed educational
specialties.
Like other contributors to efficiency, specialization solves problems for
workers as well as for their organizations. In particular, specialization per-
mits street-level bureaucrats to reduce the strain that would otherwise com-
plicate their work situation. A lawyer in a law reform unit need not balance
the demands of incessant case-load pressures, while his or her colleague who
has high case-load assignments is relieved from considering the larger issues
that clients' cases present. The social worker concerned with eligibility is re-
lieved of concerns for clients' social integration, while the income mainte-
nance worker need not worry whether clients receive undeserved support.
It is undoubtedly appropriate for some workers to be trained in areas that
others are not trained in. Not every teacher, for example, need know French
or Hebrew or Chinese for schools to provide training in languages other than
English. But some specialization relieves other workers from developing
skills they should have. As I have suggested, community relations specialists
relieve others of responsibility for concern with treatment of minorities.
Special community advocates may function to relieve others of responsibility
for being advocates themselves. Even the case of language specialization is
The Client-Processing Mentality
not so obvious as it might first appear. For should not all teachers in some
city schools know Spanish to be able to converse with a large proportion of
their students? Why should the Spanish teachers and the teachers of His-
panic background have responsibility for communicating with Spanish-
speaking students? Specialization in this case relieves the other teachers of
an important complication in their work lives.
Specialization permits street-level bureaucrats to avoid seeing their work
as a whole. Once specialized they are expected, and expect themselves, to
pursue an agenda that calls for the deployment of a restricted set of (perhaps
highly developed) skills toward the achievement of a result defined by those
skills. Specialists tend to perceive the client and his or her problems in terms
of the methodologies and previously established processing categories that
their training dictates. 12 Rare is the specialist who retains a comprehensive
conception of the client and the alternatives available for processing. In
some fields, such as special education, critics have advocated the training of
general specialists capable of working with children with any learning dis-
ability or physical or psychological behavior. (This confirms the obvious:
teachers should be well trained for the job, and the base of practice and
theory from which they should operate has expanded significantly.)
Public institutions generally have conflicting or ambiguous goals for good
reason. They embrace ambiguity, contradictions, and complexity because
the society is unable and unwilling to abandon certain fundamental aspira-
tions and expectations in providing public services. Specialists undoubtedly
bring important skills and orientations to organizations that cannot develop
them in their staff as a whole. Yet specialization and task specificity should
be analyzed to discover those circumstances in which the costs of relieving
street-level bureaucrats from contradictions and ambiguities may be higher
than the benefits.
147
PATTERNS OF PRACTICE
what kind of institution they will run. Thus hiring becomes more rational
because objectives are clearer, and employees have a clearer sense of what
they are expected to achieve.
In recent years considerable attention has been devoted to the trend
towards "medicalization" of social problems. Advanced by physicians and
supported by a public anxious to think that there are "solutions" to behav-
ioral "problems," the medical model has intruded into the worlds of educa-
tion and corrections, and other environments in which human development
is at issue. This trend has been correctly understood as undermining the po-
litical and social status of individuals, who, labeled "diseased" or "sick," are
expected by the society to accept others' definitions of their circumstances
and means for recovery. The significance for social control is substantial.
What in other times might be understood as rebellious behavior may now be
processed as mere sickness, implying no indictment and certainly no cul-
pability on the part of social institutions that may have contributed to the
genesis of the behavior.
Why has the medical orientation become so prominent? The influence of
physicians and the high regard in which most people hold them surely pro-
vides part of the answer. But this does not fully explain the attraction of the
medical orientation to say, educators, who in some respects have competing
professional perspectives.
A substantial addition to understanding the attraction of the medical mi-
lieu in education, corrections, and other fields may be gained by recognizing
the ways in which the introduction of a therapeutic milieu contributes to
simplifying the goal orientations of public service workers. It provides a
defense against personal responsibility of the worker by resting responsi-
bility for clients in their physical or psychological development. It provides a
theory of client behavior to help explain the complex world of the street-
level bureaucrat. And it provides a clear statement of clients' problems in
terms of which responses can be formulated. The hegemony of the medical
model may be explained not only by the influence of physicians but also by
the way it helps street-level bureaucrats solve problems of goal complexity.
This is not to say that goal clarification and reconstruction of work objec-
tives have no value. Schools that assert that reading is primary may be able
to achieve results that elude schools with more diffuse goals. There are un-
doubtedly physiological dimensions to deviant behavior in some instances,
although the pharmacological cure is sometimes worse than the disease. The
question is whether or not public institutions make their objectives and ori-
entations manifest and the costs of their choices clear, and whether or not it
is appropriate to abandon some goals or concentrate more on others.
The Client-Processing Mentality
DEFENSES AGAINST DISCRETION
Street-level bureaucrats sometimes cope with their jobs by privately mod-
ifying the scope of their authority. Imposing restrictions on the scope of
their powers frees street-level bureaucrats from perceived responsibility for
outcomes and reduces the strain between resources and obj ...ctives.
Denying discretion is a common way !o limit responsibility. Workers seek
to deny that they have _influence, are free to make decisions, or offer service
alternatives. Strict adherence to rules, and refusals to make exceptions when
exceptions might be made, provide workers with defenses against the possi-
bility that they might be able to act more as clients would wish. "That's the
way things are," "It's the law," and similar rationalizations not only protect
workers from client pressures, but also protect them from confronting their
own shortcomings as participants in public service work. 15 At times these as-
sertions are best understood as strategies to deflect clients' claims. But at
other times they are best understood as rigidly held attitudes that partially
have their origins in, and are bolstered by, distress over the gap between ex-
pectations and perceived capability.
Agencies often impose rigidities on their workers. For example, in the late
1g6os, when the welfare rights movement began to pressure welfare workers
to make discretionary grants to large numbers of recipients, welfare depart-
ments throughout the country eliminated discretionary special-grant awards
for furniture and other items. Thus the departments removed from workers
a discretionary option. This circumscribed their power but also eliminated
the tension between the workers' desires to help clients and their need to
control disbursements.
Another way in which agencies help solve employees' role tensions is by
extensively promulgating rules specifying official procedures. From the
point of view of reducing role tensions it is less important that rules are not
necessarily followed than that they are available as authoritative materials
with which street-level bureaucrats can renovate job conceptions to better fit
work realities. Thus rules not only order work but also function to order
workers' role conceptions. 16
Earlier chapters have focused attention on street-level bureaucrats' devel-
opment of work routines to process clients and otherwise treat their respon-
sibilities. These routines often represent more than mere instruments of ef-
ficiency. Street-level bureaucrats also develop attachments to modes of
practice. They appear to feel that their jobs require the routines. In some
street-level bureaucracies,. routines of practice become so dominant that
workers seek to negotiate the routines rather than to obtain the objective for
which routines were presumably developed.
149
PATTERNS OF PRACTICE
ter defend against the assaults to the ego which the structure of street-level
work normally delivers. The teacher's pet is not only an obedient child but
also one who confirms to the teacher the teacher's own capability.
There is another important reason to consider street-level bureaucrats'
conceptual modifications of the clientele. Just as differentiation of clients
supports rationing and other practices of organizing work, it also supports
private modifications of conceptions of work. Conceptions of the job imply
conceptions of the clientele. One cannot practice without an implicit model
of the people on whom one is practicing. An open classroom demands a con-
ception of children as requiring relatively greater freedom and flexibility
than are available in a traditional classroom. A psychiatrically oriented drug
center is founded on a different model of human motivation than a center
organized around peer interaction and self-help.
Street-level bureaucrats who are unable to provide all clients with their
best efforts develop conceptual mechanisms to divide up the client popula-
tion and rationalize the division. The differentiation of clients discussed in
previous chapters thus not only provides a rationale for allocating scarce
resources, but it also serves to help street-level bureaucrats justify their jobs
to themselves. The frequency with which street-level bureaucrats are ob-
served to divide up the client world conceptually suggests the importance of
this dimension of work in sustaining street-level practice.
The psychological importance of private reconceptions of the clientele can
be traced in the primary divisions of the client world. For example, unsanc-
tioned distinctions between worthy and unworthy clients narrow the range
of clients for whom street-level bureaucrats must provide their best efforts.
Street-level bureaucrats often respond more favorably to clients who are
helpful or cooperative in their own treatment, or who appear to be particu-
larly responsive to help. Orienting services toward cooperative clients, or
clients who respond to treatment, allows street-level bureaucrats to believe
that they are optimizing their use of resources. At the same time these per-
ceptions help condone service denials (or even routine treatment) by per-
mitting the private judgment that some clients absorb more than their fair
share of resources.
Perhaps the most familiar syndrome of private reconceptions of clients
concerns locating responsibility for client difficulties. Assumptions about
who or what is responsible for clients' situations are significant conceptual in-
struments by which street-level bureaucrats distance themselves from
clients. For example, the tendency of helping professionals to blame the vic-
tim, attributing the cause of clients' situations to the individuals themselves
The Client-Processing Mentality
without considering the role of social and environmental contexts, locates re-
sponsibility in a place that absolves the helper from blame. 20
There are many examples of blaming the victim. Chronically unemployed
men are described as shiftless and unwilling to work when their situations
might be attributed to the structure of employment and previous job avail-
ability. Students' learning difficulties are explained by focusing on their lack of
motivation rather than on the skills of the teachers and the atmosphere of the
school. Blaming clients for failing to keep appointments protects street-level
bureaucrats from the possibility that prior interviews have discouraged or
alienated them. Instances of teachers beating children who clearly display
signs of mental disturbance provide particularly brutal illustrations of the ap-
parent need of at least some street-level bureaucrats to attribute self-direc-
tion to noncompliant clients. 21 If the client is to blame, street-level bureau-
crats are shielded from having to confront their own failures or the failures of
the agencies for which they work.
An opposite but functionally equivalent mode of perceiving clients also
serves to absolve street-level bureaucrats from responsibility for service fail-
ures. This is the tendency to take an entirely environmental point of view
and perceive clients exclusively as the products of inadequate background
conditioning. Thus if children are perceived as primitive, racially inferior, or
culturally deprived, teachers can hardly fault themselves if their charges fail
to progress. 22 Similarly, job training counselors who explain failures by
clients' low motivation stemming from the discouragement experienced by
ghetto youth can avoid dealing with their own failures to make the program
meaningful.
Undeniably, there are cultural and social factors that affect client perfor-
mance, just as there is a sense in which people are responsible for their ac-
tions. However, it is important to note that these explanations function as
cognitive shields, reducing what responsibility and accountability may exist
in the role expectations of street-level bureaucrats. Moreover, because these
explanations of responsibility are illegitimate in terms of formal agency pol-
icy, they remain beneath the surface, unstated. Thus when they implicitly
form the basis for decisions about clients they contribute to misunder-
standing and to the resulting hostility of clients toward the agencies acting
upon them.
Given the imbalance in power between clients and their agencies, not all
clients will respond with hostility to decisions based on these implicit as-
sumptions. Perhaps more commonly, clients accept the implicit assumptions
of responsibility; then these conceptual structures contribute to client com-
153
PATTERNS OF PRACTICE
pliance with agency policy. Clients may accept responsibility for their cir-
cumstances without reference to the environmental conditions that they ex-
perience. Or they may regard their situation as hopeless because their
environment is so antagonistic to improvement. Each attitudinal set works
against personal movement and growth. 2 3
This is not to say that one can easily strike out for one explanation of re-
sponsibility over another. Structural explanations of clients' circumstances
are important in order to direct attention to changing the political, eco-
nomic, and social structures that circumscribe and dictate the possibilities of
action. For if environmental factors do not prescribe life changes they cer-
tainly structure the range of opportunities.
Similarly, in important respects clients to some degree must be responsi-
ble for themselves. Without this assumption there can be no client growth
within the current structure of arrangements and no client contribution to
changing those arrangements, individually or collectively. Erving Coffman's
insight into the relationship of client responsibility to absolving explanations,
developed in his study of prisons, mental hospitals, and other "total" institu-
tions, has generally wider applicability.
Although there is a psychiatric \iew of mental disorder and an environmental Yiew of
crime and counterreYolutionary actiYity, both tieeing the offender from moral re-
sponsibility for his oflense, total institutions can little afiord this particular kind of de-
terminism. Inmates must be caused to self-direct themselves in a manageable way,
and, for this to be promoted, both desired and undesired conduct must be defined as
springing fiom the personal will and character of the indi,idual inmate himself: and
defined as something he himself can do something about. 24
154
The Client-Processing Mentality
Street-level bureaucrats hold private views that affect the distribution and
quality of services, and they hold these views intensely. Their biases, when
they exist, are difficult to interrupt. Why should this be so when street-level
bureaucrats, more than most people, have regular opportunities to discon-
firm stereotypes?
A partial possible explanation has already been suggested. First, segmen-
tation of the client population complements work practices that are them-
selves compromises, and it also complements the resulting reconceptions of
work objectives. In other words, patterns of practice, conceptions of the job,
and conceptions of the clients must fit together if street-level bureaucrats are
to resolve work contradictions successfully. Private conceptions of the clien-
tele will be developed in proportion to the need to come to a private resolu-
tion of the contradictions in the work.
Second, conceptual modifications of the clientele tend to accept and build
upon general social attitudes, and thus are reinforced in everyday life. Fa-
voring clients who are underdogs or discriminating against clients consid-
ered socially unworthy may partly be explained by the sympathies and an-
tipathies of the general society. Sociologist Howard Becker reports that
children may be morally unacceptable to teachers in terms of values cen-
tered around health and cleanliness, sex and aggression, ambition and work,
and age-group relations. These considerations are particularly likely to be
salient when class discrepancies between teachers and pupils are signifi-
cant. 26 These responses to childrens' characteristics are not likely to be
unique to teachers. But when teachers do respond to children in these
terms, their responses have implications for public policy.
Other conceptions of clients appear to enhance feelings about job ac-
complishments even when they seem to run counter to prevailing social
norms. Consider the case of social service workers who would rather be as-
signed to child abuse than to child neglect cases. Although child abuse is a
particularly unattractive crime the anomaly appears to be explained by the
greater likelihood that child abuse cases will respond to intervention, while
the typically passive child neglector is less likely to respond to social work-
ers' assistance. 27 It would appear that clientele segmentation is usually con-
sistent with prevailing social norms, but it is not wholly explained by them.
Third, various aspects of the ways in which street-level bureaucrats re-
ceive information about their work contribute to conceptual modifications of
the clientele. Illustrative validation, self-fulfilling prophecies, rational-
izations that excuse failure, and selective retention of information tend to
confirm rather than disconfirm workers' attitudes about clients.
Finally, street-level bureaucrats work in a milieu in which their co-
155
PATTERNS OF PRACTICE
workers have similar needs to segment the client population. Thus attitudes
prejudicial or beneficial to certain clients are likely to reverberate among,
rather than be contradicted by, other workers.
Street-level bureaucrats have a need to modify their conceptions of clients
quite apart from but usually consistent with the prejudices of the general so-
ciety. And they work in a structure that tends to confirm the validity of their
biases. The general argument of this section, based on observations that
street-level bureaucrats consistently introduce unsanctioned biases into
client processing, suggests that it would be difficult to eliminate client dif-
ferentiation without changing the structure of work for which these biases
are functional.
This is not to say that any particular bias is necessary to cope with the
work. No doubt classes of clients may be treated in markedly different ways
if administrators pay enough attention to specific behavior of workers. But
without changes in the work structure one ought to expect that biases will
soon develop in other areas, or that the old biases will soon emerge in new
forms in the absence of considerable vigilance.
PART IV
THE FUTURE OF
STREET- LEVEL
BUREAUCRACY
CHAPTER 11
159
THE FUTURE
160
The Assault on Human Services
cies. Whatever theoretical advantages exist in other approaches, efforts to
obtain bureaucratic accountability will have the greatest immediate impact
on workers and clients alike.
To utilize organizational incentives and sanctions, at least the following
conditions must prevail. These conditions are the prerequisites of a bureau-
cratic accountability policy.
1. Agencies must know what they want workers to do. Where the objectives are
multiple and conflicting, agencies must be able to rank their preferences.
z. Agencies must know how to measure workers performance.
3 Agencies must be able to compare workers to one another to establish a stan-
dard for judgment.
4 Agencies must have incentives and sanctions capable of disciplining workers.
They must be able to prevail over other incentives and sanctions that may
operate.
forts simply generally ineffective? There are several respects in which con-
trol practices can actively subvert service quality.
First, specification of methods of client treatment under the guise of ob-
taining accountability may actually result in reductions in client services.
There is often a thin line between inducing workers to better conform to
agency policies and inducing workers to be open to fewer options and oppor-
tunities for clients. For example, during the Nixon Administration the De-
partment of Health, Education, and Welfare attempted to increase welfare
employees' accountability by auditing their error rate in accepting clients for
welfare. This policy reduced services by providing incentives for welfare
workers only to reduce errors that favored clients. Federal guidelines did
not then call for reducing error rates for the potential welfare population as a
whole. If it had done so the applications of all welfare applicants would have
been audited, both those accepted and those rejected. Scrutiny of welfare
workers' decisions strictly in terms of whether or not they were too lenient
amounts to narrowing the role of welfare workers, reducing their account-
ability to clients and to professional standards of conduct.
Second, supervision of subordinates with broad discretion and responsi-
bilities requires assertions of priorities in attempting to increase account-
ability. Police departments may scrutinize traffic tickets, vice arrests, or in-
terracial encounters between police and citizens. But they cannot
meaningfully hold officers accountable for everything all the time. If every-
thing is scrutinized, nothing is scrutinized. Thus efforts to control street-
level bureaucrats not only affect those areas that are management targets,
but they also affect those areas that are not the focus of management efforts,
since by implication those efforts will not come up for surveillance. Efforts to
increase accountability in some areas may come to be regarded as the only
areas in which accountability will be sought and behavior scrutinized.
Third, many management control efforts provide a veneer of account-
ability without in fact constraining behavior very much. Management con-
trol systems have symbolic value, providing concerned publics with reassur-
ances that employees are accountable even when they are not. Introduction
of management systems at least temporarily permits agencies to deflect criti-
cisms as citizens find it very difficult to challenge the emperor who officials
say is fully clothed, appearances and personal experiences to the contrary
notwithstanding. 6
GOAL CLARIFICATION
One of the conspicuous features of many public services is the ambiguity
and multiplicity of objectives. How can accountability be achieved, ask the
The Assault on Human Services
critics, if public officials are unclear about their objectives? The desirability
of clarifying (and then putting into operation) agency objectives to increase
accountability stems from the force of this observation and recognition that
a bureaucratic accountability policy requires specification of objectives (as
suggested above).
Surely it is desirable to clarify objectives if they are needlessly and irrele-
vantly fuzzy or cgntradictory. Surely it is easier to run an effective agency if
you know what you are supposed to be doing. However, while agency goals
may be unclear or contradictory for reasons of neglect and historical inertia,
they may also be unclear or contradictory because they reflect the contra-
dictory impulses of the society the agency serves. Schools attempt to in-
struct, but they also inculcate attitudes toward social behavior and c'itizen.-
ship. They do this not because educators are fuzzy but because both these
objectives are favored by parents (and because there is no convincing case
that they are mutually incompatible). Criminal justice institutions are
oriented toward punishment and rehabilitation not because judges and cor-
rections officials are simpleminded but because the society has impulses
toward reforming as well as deterring criminals.
The public service areas of education, corrections, and welfare in recent
years have all been subject to efforts to increase accountability through goal
clarification. Educators have sought to concentrate on reading to the exclu-
sion of other educational objectives; corrections analysts have sought to build
up the role of punishment and make it more certain at the expense of empha-
sis on rehabilitation; welfare reformers have successfully separated decisions
on income support from social service provision. The dilemma for account-
ability is to know when goal clarification is desirable because continued am-
bivalence and contradiction are unproductive, and when it will result in a
reduction in the scope and mission of public services. The problem of goal
ambiguity has contributed to the discrediting of institutions providing ser-
vices in social work, corrections, and mental health, and to the dismantling
of many programs to provide assistance in these areas. But it requires the
most serious inquiry to determine the long-term implications of requiring
the former and potential clients of these institutions to have recourse exclu-
sively to noninstitutional community and personal resources.
PERFORMANCE MEASURES
The development of performance measures is critical to a bureaucratic ac-
countability policy. Administrators make great efforts to develop perfor-
mance measures in order to control employees' behavior.
THE FUTURE
168
The Assault on Human Services
Thus teachers resist being measured by the progress of their pupils unless
adequate provision is made to control for their students' previous levels of
achievement (and, more important) for their students' capacity to learn.
Thus police officers would object to utilizing arrests per capita per available
officer as measure of performance unless controls were introduced for the
propensity of criminal behavior in the district. Comparing districts by out-
come tends to be useless because of the inadequacy of such controls.
Some advocate that measures such as these be deployed in order to dis-
cover deviations from normal practice, so that workers who deviate from the
norm can be brought into line. These proposals have some merit, but here
the problem is that unless one is confident that the best workers or dis-
tricts are doing a good job, such comparisons may simply institutionalize
mediocrity.
Recent attention has also focused on utilizing surveys of client satisfaction
to obtain information on workers' performance. Resistance to such surveys
arises out of professional skepticism that performance that pleases clients
may not be related to high-quality practice. The popular teacher may not be
the most effective instructor; a popular judge may not be the fairest judge.
Still, in a comprehensive assessment program, client satisfaction surveys
have their place if client satisfaction is indeed deemed desirable (as it usually
is).
Street-level bureaucrats' interactions with clients tend to take place in
private or beyond the scrutiny of supervisors. Interviews are held in private
offices and under norms of confidentiality. Teaching is done in classrooms
that principals and supervisors do not normally enter; if they do, they provide
notice so that the teaching, like a performance, may be changed by the pres-
ence of an audience. Police officers, although they do take action in public,
normally do so without observation by other officers or supervisors. The ex-
ception is the officer's partner, who is compelled by police norms to shield
his partner from criticism. Of the street-level bureaucrats we have studied
only judges tend to make their important decisions in public.
This fact provides a barrier to an important potential source of perfor-
mance measurement. It might be possible for street-level bureaucrats to
scrutinize each others' work and provide assessments of quality. But given
the structure of these agencies such scrutiny would be highly obtrusive in
relations between workers and clients and very costly if engaged in on a
widespread basis. Thus public service agencies rarely engage in direct obser-
vation of their line workers, but depend upon the written record supplied by
their workers (the reliability of which was discussed earlier).
169
Accountability and Productivity
Thus far I have focused discussion on some of the major difficulties in devel-
oping an administrative accountability policy. But are there any negative ef-
fects of such policies? For example, what is the harm of attempting to de-
velop performance measures? It may be difficult to measure performance,
but perhaps we are simply at the beginning of the development of a manage-
ment tool. Perhaps the current measures of performance are not entirely ad-
equate, but they may have their uses, and they may be increasingly refined.
This is the rhetoric of those who are committed to achieving bureaucratic
accountability and recognize the inadequacy of current measures, but who
apparently have faith that their approach is ultimately oorrect. 14 This line of
discussion would have us believe that there are some benefits to current ef-
forts to develop accountability through improved performance measure-
ment, that these benefits are likely to increase, and that there are no signifi-
cant costs. Surely management benefits from operationalizing and
attempting to develop measures of worker performance. Even if the pre-
ferred behavior cannot be adequately measured, performance measurement
and monitoring can signal workers powerfully concerning which aspects of
performance are most salient.
However, in the current period bureaucratic accountability policies also
have negative consequences because of the competing demands on, and of,
administrators. Currently, public agencies are under enormous pressures to
minimize costs and increase productivity. They are under pressure to reduce
government expenditures or keep them from rising. They are under pres-
sure to increase productivity in order to maintain services, or claim that they
are maintaining them, in the face of financial stringency. And they are under
pressure to increase productivity in order to justify employee pay increases,
which they are under particular pressure to grant because of the impact of
inflation on wages. (The only way to stabilize government budgets when ser-
vices cannot be reduced beyond a certain level and costs are rising is to
increase the productivity of the present work force. Organized workers
argue 'that they have no incentive to increase productivity unless they can
share in the gains made because they work harder and cooperate with the
reorganization of work often entailed by productivity reforms.) 15
Productivity in the public sector summarizes the relationship between the
utilization of resources and the resulting public services product. Productiv-
ity may improve in three ways: when costs remain the same while public ser-
vices increase; when costs decline while services remain the same; or when
The Assault on Human Services
costs increase but services increase still more. Schematically, there are two
dimensions to public services implied here-<>ne qualitative, the other
quantitative. If the quantity increases or remains the same, but services have
declined qualitatively, productivity increases have not necessarily taken
place. 16 If more garbage is picked up on the streets without increasing the
crews, but half of it is strewn back on the streets, productivity has not
increased. The debasement of service is what infuriated New Yorkers when
transit workers were given an increase in pay based upon alleged gains in
productivity. It appeared, however, that the Transit Authority had been able
to provide services with fewer personnel only by increasing the time be-
tween trains and reducing the number of cars in operation. 17 In this view the
transit workers had been falsely credited with improved productivity.
These are the essential elements of productivity. In practice, however,
debasement of services is rarely taken into account, although the problem is
given lip service by productivity theorists. This may be true for several
reasons. First, if the quality of service is difficult to measure, so is reduction
in service quality difficult to measure. Second, there are many ways to save
money by eroding the quality of service without appearing to do so. They
include offering services on a group rather than an individual basis, substi-
tuting paraprofessionals, often paid from other sources, for regular staff, and,
conversely, forcing professionals to handle clerical and other routine chores,
reducing the time they have to interact with their clients. 18 Additionally,
street-level bureaucrats can narrow the range of situations in which they will
act. Examples include legal services offices that decide to take only
emergency cases, police departments that decide to neglect selected infrac-
tions, and schools that offer reduced programs. Each of these techniques
permit managers tp give the appearance of maintaining services while reduc-
ing costs.
Third, in the current period pressures experienced by public managers to
reduce the budget and improve productivity are pressing and general, while
the constituency for maintaining service quality is disorganized, weak, or
nonexistent. Only clients experience service quality reduction, and they are
severely constrained in comparing their experiences with others and organ-
izing collectively to oppose service quality debasement. Ironically, the
greatest opponents of service quality debasement are street-level bureau-
crats themselves. For them debasement often means harder work, less job
satisfaction, and greater individual problems with clients. Yet they are cross-
pressured by the interest they have in helping their agencies appear finan-
cially responsible and by their collusion with public officials to share finan-
cially in productivity increases.
THE FUTURE
But there is more to the debasement of public services than pressures and
interests. A large part of the problem stems from the orientation toward
measurement, precision, and scientific management itself. Consider the
formula productivity =service quantity and quality/cost. Two of the deter-
minants of productivity, service quantity and cost, are easy to measure; the
third, service quality, is virtually impossible to measure. Managers under
pressure to improve productivity are likely to try to cut personnel or obtain
more work from existing personnel because these are the terms of the eq ua-
tion for which measures are available and which managers can manipulate.
Thus staffs are reduced to bare bones without reduction in responsibilities.
Thus staffs are asked to do more without increases in personnel. 19
There is always an implicit tension between resource constraints and the in-
exorable demands for public service. However, this tension is rarely mani-
fest politically. The budgets (and employment rolls) of street-level bureau-
cracies rise not only with increases in population to be served, but also with
higher standards of what citizens are entitled to, a decline in the availability of
comparable private services, and the perceived need for more effective and
improved agencies of social control, coercive or manipulative. The impulses
to increase expenditures in these areas rarely have been challenged in terms
of resource constraints. In the period since World War II federal govern-
ment subsidies to state and local public services have postponed or softened
the confrontation between revenues and expenditures in areas such as public
health, education, police, and welfare. However, the current period, charac-
terized as a fiscal crisis among state and local governments, forces recogni-
tion of the relationship between what people get from government and what
jurisdictions are willing to pay.
At best the term "fiscal crisis" is reserved for situations in which financial
agreements and long-standing patterns of practice can no longer be honored,
as when a political jurisdiction cannot meet its payroll or honor its commit-
ments to lenders. But the term is also used much more loosely to mobilize
people to believe that there is something wrong or that there is a problem as-
sociated with current and projected expenditures relative to available reve-
The Assault on Human Services
nues and other income. If political and economic elites are successful in
promulgating a sense of crisis they are able to make manifest and set the
terms of confrontation between governmental expenditures and income. If
in other times social services, for example, grew in response to perceived
societal needs, in a fiscal crisis the imperatives for service development are
subordinated to the demands of perceived revenue limitations. 20
Like other political confrontations, the management of fiscal crises has re-
distributive consequences. The costs of responding to the needs of expendi-
ture constraints do not fall evenly or randomly on the population as a whole,
but rather affect different segments of the population differentially. The fis-
cal crises of the cities provide a focus and an apparently benign rationale for
attacking and injuring the provision of public services. And they demon-
strate the vulnerability to attack of high levels of public service quality.
The fiscal crisis of the cities affects the quality of service delivery in two
significant ways. First, services are rationed in various ways, maintaining the
appearance of service while reducing and debasing it in practice. This is not
to say that legitimate savings cannot be realized by eliminating "real" waste
and duplication that may exist. 21 However, in city administration these
"real" savings tend to be concentrated in areas in which questions of re-
source deployment are paramount, not in areas in which the provision and
nurturing of interactions with street-level bureaucrats is at issue. It is con-
spicuous, for example, that when administrators take pride in productivity
savings it is in the sanitation department that the greatest successes are often
realized. In police departments dispatch (deployment), not interactions with
citizens, is the area of concentration in productivity campaigns. Most street-
level bureaucracies have to contend with impulses to reduce the amount of
time workers spend with clients, not the reverse.
For public officials the problem of managing the fiscal crisis consists of
reducing expenditures while minimizing the apparent impact of the cuts.
This is why cuts will initially be said to eliminate waste and duplication
whether or not waste exists or duplication takes place. Rationing typically
means increasing the costs to clients of seeking services, while maintaining
the service shell, or reducing services to decrease potential benefits. Both
prospects are likely to lead to lower client demand. Closing neighborhood
branch offices while continuing to offer services from a central (downtown)
location is a typical technique for achieving this. Reducing the number of
telephones or receptionists reduces the number of inquiries. Increasing the
response time in investigating a complaint 1educes the efficacy of complain-
ing and hence future volume. City agencies often experience decreasing
173
THE FUTURE
ability to process citizen needs at the same time that they are experimenting
with techniques to improve their performance. The public message regard-
ing agency responsiveness becomes mixed, to say the least.
When public managers decide to fight demands for service reduction they
say that all waste, duplication, and nonessential services have been cut, and
that any further cuts will be in essential services. Their ability to make this
claim depends on general public perception of the importance of the agency
and public employees' collective capacity to resist. Thus schools can make
the claim more effectively than welfare agencies, police departments more
effectively than sanitation departments. Cuts in service provision obviously
may affect service quality, but is is impossible to determine from public rhet-
oric where the politics of distributing urban resources to public employees
ends and injury to service delivery begins.
A second dimension of response to the fiscal crisis is personnel reductions.
Personnel practices are particularly important because salaries comprise the
bulk of urban budgets; thus savings must be sought in the area of public
employment.
Personnel practices in the fiscal crisis tend to follow a path of severity. Ad-
ministrators first make it more difficult for agencies to replace workers who
leave. Next administrators suspend hiring, then slow wage increases or
freeze wages, then begin to lay off workers. All of these steps have implica-
tions for service provision but the most important point is that each of these
steps reflects increasing penalties to public employees, regardless of the
implications for clients. Priorities are set by imperatives of labor-
management relations, not by the needs of clients for service provision.
Increasing the difficulty of replacing workers and freezing employee rolls
represent managers' efforts to realize savings through attrition. They at-
tempt to reduce personnel rolls by not replacing those people who retire or
otherwise leave their jobs. Since in normal circumstances the rate of exit in
almost any line of work is substantial, a significant reduction in employment
can be realized over several years without firing anyone. Realizing savings
through attrition accomodates public managers' needs to keep peace with
organized public employees, but it creates substantial costs for service provi-
sion above and beyond the obvious reduction in force. Since workers do not
retire or otherwise leave their jobs in response to agency priorities, the in-
cidence of turnover is unevenly distributed in the work force. This means
that important gaps in service provision occur. Workers important to the
operation oi a particular office, or those who possess critical skills, may be
the ones to quit rather than the most marginal employees. If the critical posi-
174
The Assault on Human Services
tions they vacate are left unfilled, the injury to service provision is obvious.
But even if they are filled by employees who remain they are unlikely to be
filled well.
In the street-level bureaucracies with a high level of job specialization the
vacancies will be filled by employees who lack the required skills or re-
sources. Fifth-grade teachers will be assigned to kindergarten classes, and
physical education instructors will become math teachers when there are ex-
cesses of the former and need for the latter. But even in street-level bureau-
cracies with low levels of specialization it will be difficult to fill the vacancies
adequately. Police departments, for example, assign desk officers to active
patrol in order to maintain the patrol force when additional officers cannot be
hired. But those officers assigned to desk jobs may be more suitable for
sedentary positions than the officers they replace. Moreover, except for re-
tirees the people whp leave public service work in the fiscal crisis will tend to
be the more employable, so that the work force at the same time becomes
mediocre through attrition.
The squeeze of the fiscal crisis is particularly tight because in many ways it
reflects national rather than local conditions. If a city were financially
strapped in an otherwise healthy economy, young public employees would
be able to find work in other locations. However, all governments more or
less experience pressure to hold the line on costs at the same time, since the
fiscal crisis is significantly a function of national economic and political
trends. Thus no longer are there jobs waiting for young teachers or social
workers in the suburbs, or California, or elsewhere, as in the past.
Street-level bureaucrats' performance is not tied directly to wage incen-
tives. Promotions and raises, when they are given out, do not depend so
much on job performance as on personal relations, additional outside train-
ing, work-load handling, and other factors unrelated to client servicing.
Moreover, promotions to positions of greater responsibility are rare, since
the job hierarchies in most street-level bureaucracies schematically resem-
ble relatively flat pyramids, with existing jobs undifferentiated at the bottom
of the scale. When most teachers can look forward only to being teachers,
most police officers only to being police officers, incentives to high-quality
service may flag unless specially encouraged. 22
For these reasons it is probably incorrect to argue that wage freezes di-
rectly affect workers' motivation. Their impact is likely to be somewhat dif-
ferent. First, wage freezes and slowdowns affect the likelihood that workers
will stay in their jobs, and they force people out of jobs without firing them.
Again this affects the distribution of age and skills in street-level bureau-
1 75
THE FUTURE
cracies. Older workers will stay to protect and build pension benefits.
Younger workers will be more likely to leave public employment, somewhat
tempered by the availability of jobs elsewhere.
In periods of high inflation wage freezes effectively reduce workers' real
income as well as their income relative to workers in other sectors. The feel-
ings of deprivation that may result from diminishing workers' income in
these ways are quite different from attitudes toward wages that street-level
bureaucrats may have had in other periods. While wages may not play a
major role in creating incentives to performance, effectively reducing wages
may have a significant impact on street-level bureaucrats' attitudes toward
their jobs.
Street-level bureaucrats may be accountable to managers and clients, as
mentioned above. But they are accountable to clients on behalf of their
agencies and the public purposes they represent. To reduce workers' wages is
to shred these bonds of accountability by bringing to the surface aspects of
the wage relationship that otherwise remain obscured by the claims and
ideology of professional status and attitudes. People who accept the rela-
tively fixed civil-service formulas for wage increases will not easily accept
receiving less than they did receive, particularly as the motivation for street-
level work leads some employees to regard their work as voluntaristic to
some degree, that is, undercompensated to begin with. As wages are effec-
tively reduced street-level bureaucrats may be expected to look more to
their benefits and remuneration and less to the service dimensions of their
job.
Decruitment practices, of which firing workers is the most drastic, have
similar implications. 23 When workers are not replaced their responsibilities
are distributed among those who remain, usually without reducing the re-
sponsibilities they already have. Increasing the responsibilities of other staff
without increased compensation or work resources is the white collar equiv-
alent of the onerous assembly line speedup.
In this connection consider the modest complaint of a New York City
school ,teacher over staff reductions, increased responsibilities, and reduced
resources.
I have never been quite sure what ["increased productivity"] means exactly. How-
ever, if it means what I think it means, they would like to see us work harder than we
ever did before. If this is so, then all the proponents of "increased productivity" will
be delighted to know that we are doing remarkably well in that department.
For example, we have official classes of 45 or more youngsters and ten minutes in
which to take attendance, read circulars, distribute notices, make reports (in dupli-
cate yet), answer questions, etc. In many cases we have classes which have rosters of
The Assault on Human Services
49 or more children with 30 chairs in the room or typing classes of 47 with 3Z type-
writers. Add to this emergency coverages of classes, cafeteria patrol or other building
assignments, program problems, shortage of supplies and equipment and much
more-all of this with reduced staff-not to mention the mounds of work we take
home with us. The pressures under which we work can never be understood by any-
one who is not involved in day-to-day school activity. Yes, we have indeed increased
productivity, but in so doing we have decreased our effectiveness as human beings to
our students, our families and ourselves. 24
177
THE FUTURE
1 7Y
CHAPTER 12
180
The Broader Context of Bureaucratic Relations
advanced society is the extent and persistence of different subcultures and
classes, particularly as expressed in clients' preparation and readiness for the
impersonalism, hierarchy, and institutionalization of bureaucracy. 4 In this
connection bureaucrat-client relations in the United States may be said to
reinforce and be structured by the American system of persistent and cumu-
lative inequalities experienced by subordinate groups, particularly as this
system consigns people to poverty and low expectations of mobility, monop-
olizes the service fun.ction, and grudgingly provides for public services.
In stating this one must go beyond observing that the character of client
treatment at the hands of street-level bureaucrats reflects and reinforces
social class and racial divisions. While some may be tempted to view the
character of U.S. social services as an expression of racism and of attitudes
toward the individual in mass sodety, we cannot conscientiously stop there.
This view does not explain the development of such an elaborate service and
control apparatus. Nor does it account for the opportunities clients have to
redress grievances, find support within the service network, and resist dehu-
manized services. Nor does it account for or comprehend the wide range of
forms and structures affecting clients in street-level bureaucracies through-
out the country.
Moreover, asserting that street-level bureaucracies reinforce social cleav-
ages does not begin to account for the content ofbureaucratic behavior. Cop-
ing behaviors and adaptive attitudes may be endemic in organizational life.
But this says nothing about the nature of the coping behaviors, or the orien-
tation of adaptive attitudes. Street-level bureaucrats do not fully invent re-
sponses to work stresses but instead at least partially reflect the culture in
which their agencies are embedded. In other words, responses to work
stresses arise out of the work situation, but their content or direction are col-
ored by prevailing cultural assumptions.
In what ways do street-level bureaucracies reflect and perpetuate the val-
ues of the larger society? There are at least two respects in which the struc-
ture of relationships between workers and clients appears to be derived from
the particular character of American society.
First, street-level bureaucracies are affected by the prevailing orientations
toward the poor in the United States. These orientations include the deep
conviction that poor people at some level are responsible for the conditions
in which they find themselves, and that receiving benefits labeled "for the
poor" is shameful. These convictions are epitomized in the observation that
public programs for poor people are almost always treated in the pres:; as
costs to society, not benefits.
These attitudes toward social services for the poor amount to a general
THE FUTURE
186
The Broader Context of Bureaucratic Relations
larly, teachers who believe they are not supported by parent groups or by
school authorities (or by the public-at-large) will be unlikely to seek changes
in working conditions that go beyond those expected of normal labor-
management demands. In short, the more different groups perceive them-
selves to be in conflict with others, the more narrowly they must choose
among objectives and the less they can devote themselves to long-run goals.
A second difficulty is that the patterns of practice that develop in this work
are rooted in the fundamental coping requirements of the job. These are not
easily abandoned or changed because they are experienced by workers and
outside observers as virtual job requirements. People do not readily give up
survival mechanisms. This is one of the reasons it is easier to change articu-
lated policy from the top than to change practice from below. Policy articu-
lated from the top is not rooted in defense mechanisms developed to cope
with the job, while the policy that emerges from practice is rooted in
survival.
Curiously, a third reason it is difficult to think about changing street-level
bureaucracies is related to the appearance of flexibility and innovation.
Street-level bureaucracies are permeated with turmoil rationalized as
change-related. The social service industry of managers, management con-
sultants, public administrators, foundation officials, and academics whose
business is to tinker with social service improvement insures that public per-
ception of street-level bureaucracies is one of constant alterations in the
structure of service delivery. The volume of pilot programs, demonstration
projects, and innovations in management and personnel practices presents
the image of frequent reconstitution of public agencies. This profligacy of
reform confuses a public that cannot possibly assess the programs under-
taken in the name of reform and rarely experiences results of such reforms
directly. The pluralism of the social service industry tends to credit virtually
any kind of change so long as it has its origins in a legitimate source (which
rarely includes clients). Thus the currency of reform is debased. With so
many reform proposals no one change seems better than any other change.
Finally, thinking about significant changes in street-level practice implies
a commitment to altering or improving relations between individual workers
and clients. Yet we are profoundly shy and inexperienced in talking about
relations between and among people. We know much more about deploying
resources than about affecting working relations. It is typical for community
meetings to address issues of recruitment, procedures, incentive structures,
chains of command, and so on, instead of confronting the problems that actu-
ally brought people together in the first place-incompetent or insensitive
teachers, police officers, or social workers. It is easier to avoid these prob-
THE FUTURE
lems, the heart of community relations, and defer them as professional mat-
ters better left for professionals to handle. In any event it is difficult to
measure the quality of relationships; better to stick to dimensions of the
work more subject to administrative manipulation.
In sum, if bureaucracies mirror the society in which they develop it is dif-
ficult to change bureaucratic forms fundamentally without larger changes
taking place. While superficial changes may frequently develop in bureau-
cratic organizations-for example, changes in the relative degree of adminis-
trative centralization-in a sense there remains a deeper structure of organi-
zational relationships that are not likely to yield easily to administrative
rationalizations. This provides an additional reason for insisting that policy
implementation in street-level bureaucracy must be studied at the work
place rather than tracing policy through the bureaucratic and interorganiza-
tional systems. Not only do street-level bureaucrats exercise discretion to
such an extent that they are not easily affected by policy articulation from
above (as I have previously argued). Also, the character of worker-client and
worker-supervisor relations, no matter what the articulated organizational
policy, is likely to continue to reflect the dominant bureaucratic relations of
the society, no matter what the administrative guidelines provide.
188
The Broader Context of Bureaucratic Relations
five should be noted. First, public programs of entitlement and control pro-
vide at least the potential for mobilizing clients and sympathetic publics
toward greater accountability in implementation and administration. 16 One
of the contradictions of the American service sector is that programs con-
tributing significantly to the management and control of populations are sup-
posed to be responsive to public preferences. This, of course, does not mean
that democratic theory necessarily calls for the citizen-subjects of public
agencies to control agencies. Agencies of service and control are theoreti-
cally accountable to a larger society, of which the client population is only
one, and an often relatively powerless, constituency. Still, while public pro-
grams have many constituencies, they more or less present the potential for
formal responsiveness to client preferences. The requirements of account-
ability in theory contribute to the actual capacity of client populations to or-
ganize to change public agencies as they strive to close the gap between ac-
countability in theory and practice.
This consideration is likely to be important in proportion to the population
covered by service. One possibility is that public agencies will simply dif-
ferentiate among high- and low-status clients. But another is that service will
improve for all if high-status clients are included in the population mix. As
public-health care delivery becomes more and more generalized and less the
concern oflow-income populations, it is more likely that clients will be able
to have an impact on service quality. Likewise, parents of children in an in-
tegrated school can have a greater impact on the quality of service than can
parents of a segregated school whose needs can be more easily isolated.
The importance of this consideration increases as the public service sector
in the society grows. As people are thrown more and more into the public
sector for relief or entitlements they will be more able to make claims based
on the discrepancies between programs that would provide for them ade-
quately and programs as they actually exist. Government becomes more vul-
nerable to challenge as it assumes more responsibilities. This is why the ser-
vice state, increasingly penetrating aspects of private life, contains elements
that contradict the control it presumably exerts.
Second, professional norms of behavior toward clients provide a measure
of resistance to bureaucratization. Street-level bureauc1 ats' claims of profes-
sional status imply a commitment that clients' interests will guide them in
providing service. The implicit bargain between the professions and society
is that in exchange for self-regulation they will act in clients' interest without
regard for personal gain and without compromising their advocacy. In a
word, according to professional norms professionals' interest as expressed on
behalf of clients must be unalloyed.
189
THE FUTURE
This is not to say that street-level bureaucrats do not also confront organi-
zational demands. On the contrary, the essence of their dilemma is that they
are partly professional and partly bureaucratic. However, the potential for
appealing to the professional dimension of these work roles means that there
is an irreducible minimum consideration of the importance of respecting
clients' individuality and acting accordingly.
Even in the roles of police officers and judges, where the concepts of
client advocacy and representation might appear remote, we can see norms
of appropriate application of sanctions that in these occupations are the
primary ingredients of resistance to bureaucratic treatment of offenders. Al-
though it is tempting to see teachers, social workers, and other low-level
workers as constantly under attack for routinizing social services, one can
tum this observation around by calling attention to their resistance to bu-
reaucratization (to the extent that they have been able to resist), largely
stemming from their recognition of the importance of this aspect of profes-
sional identity. (The role of professionals in reform of street-level bureau-
cracy is discussed at greater length in the next chapter.)
Third, street-level bureaucrats by definition interact constantly with
clients. This provides the salutory condition that workers must continually
attend to the people they are supposed to serve and their problems. How-
ever elaborate the defense mechanisms developed to shield themselves from
the enormity of clients' needs, street-level bureaucrats at some level retain a
sense that the people with whom they come in contact are not sufficiently
served by the agencies designated to do so. Thus, one might speculate that
street-level bureaucrats more than other organizational workers are able to
retain a concept of the notion of need in relation to what is actually being
provided. This residual awareness may provide a resource that can be
tapped.
Fourth, lower-level workers maintain a degree of control over their work
environment. Individually street-level bureaucrats exercise discretion to
control the work situation. Collectively many street-level bureaucrats are
able to have a significant say in the rules under which they are employed.
Particularly at the individual level this discretion is not likely to be signifi-
cantly eroded so long as street-level bureaucrats' jobs require them to make
discretionary judgments that cannot be entirely programmed.
Finally, there is a distinct but neglected precedent for organized public
employees championing the needs of clients. Teachers have included limita-
tions on classroom size as an objective to be sought through collective
bargaining. They have sought this objective not only to improve working
conditions but also to create the environment in which they could function
I go
The Broader Context of Bureaucratic Relations
optimally as teachers. Likewise, social workers have struck on behalf of
improved benefit levels for clients. 17
The cynic may wish to point out the strategic advantage to public workers
of couching bargaining objectives in altruistic terms (although managers are
equally guilty, insisting that they act on behalf of taxpayers and the economic
well-being of the community). Still, cynical or not, such alliances, the stuff of
politics, may be exploited by client groups, particularly when, in collective
bargaining in the fiscal crisis, wage gains are subordinated to improvements
in working conditions.
The impulse to provide fully, openly, and responsively for citizens' service
needs exists alongside the need to restrict, control, and rationalize service
inadequacies or limitations. This is the central contradiction of social ser-
vices. It is more than simply a tension between costs and benefits. It is criti-
cal to reassure mass publics that their elemental needs will be taken care of if
they are not met privately and to rationalize service inadequacies by deflect-
ing responsibility away from government.
Through street-level bureaucracies the society organizes the control, re-
striction, and maintenance of relatively powerless groups. Antagonism is
directed toward the agents of social services and control and away from the
political forces that ultimately account for the distribution of social and mate-
rial values. Thus the American system of service delivery and control is
shaped by the aspirations of the population and by the requirements of the
larger political and social system. In this sense the United States, no less
than other political systems, lends public bureaucracy its particular
character.
CHAPTER 13
Proposals for greater client autonomy generally suffer from the fact that
clients tend to remain relatively powerless. Clients accorded greater collec-
tive influence may not possess the bureaucratic skills necessary to operate in
the policy arena, or they may inherit control over programs or facilities so
bankrupt that they defy significant management improvements. Tenant
management of underfunded and poorly maintained public housing, for ex-
ample, may fail to improve service more for financial and structural reasons
than for reasons of client capability. In this sense giving clients control over
public facilities may contribute to social control, as tenants contest with
other tenants over scarce jobs and project resources.
Nonetheless, proposals to increase client autonomy must be vigorously
studied for their potential contribution to changing street-level rela-
tionships. One approach to change in street-level bureaucracy is to eliminate
public workers as buffers between government and citizens. A class of pro-
posals utilizing such an approach is represented by plans to issue service
vouchers to citizens. By providing clients with claims on public or private
service agencies, sponsors of voucher proposals hope that agencies would be
more responsive to client preferences in order to attract their patronage.
1 93
THE FUTURE
194
Support for Human Services
inate mediating public employees will accomplish these objectives. They
may spawn quasi-public agencies that have the potential for replicating the
difficulties of the agencies they replace. They may develop entitlement, reg-
ulatory, or service bureaucracies that perpetuate bureaucratic experiences.
Home care, for example, frees people to stay out of the hospital under cer-
tain circumstances, but it still requires a bureaucracy to certify eligibility, to
promulgate and monitor standards for service providers, and to see that ser-
vice personnel are hired to provide home care. In general, questions of
supply and maintenance of standards remain in all service areas so long as
the government retains ultimate responsibility.
These alternative perspectives on service provision suggest opportunities
to define the relationship of providers and clients differently. However, they
do not fundamentally reorganize the need for service in many instances and
do not offer guidance where street-level bureaucrats remain in a controlling
relationship. A sharp need continues to provide a better balance of power
between street-level workers and clients. A better balance would be
achieved if the following developments were encouraged.
Wherever possible, opportunities should be seized to demystify street-
level bureaucracies and the practices in which they engage. Workers should
be taught how to communicate with their publics in plain language, and
clients should demand explanations they can understand. Client advocates
should be sponsored and trained to guide clients through the bureaucracy,
to obtain answers they are otherwise unable to get, and to represent clients
to workers where they would otherwise be intimidated. Guides to clients'
rights and maps of bureaucratic systems should be developed; more impor-
tant, street-level bureaucracies should simplify procedures to make service
systems more manageable without expert intervention.
Simple practices should be developed to make street-level bureaucracies
more accountable to clients. Requiring workers to provide summaries of the
transactions clients experienced but may not have fully comprehended
would be a significant step forward in some places. Routine reviews to deter-
mine whether clients were receiving all benefits to which they ~ere entitled
would place the burden of programming for clients on public employees.
Such details would modestly contribute to the development of more recipro-
cal relationships.
As a matter of public policy we should welcome investigations by public-
interest law firms, legal services offices, government agencies, and others
challenging prevailing practices where those practices entail responsible
allegations of inhumane service or systematically neglected clients' rights.
We should recognize that the discretion of street-level workers is uncer-
195
THE FUTURE
tainly monitored at best and that governments that create these bureau-
cracies may properly oversee their direction by encouraging client as well as
bureaucratic scrutiny.
The struggle of clients to organize and obtain some control over service
provision should be respected and encouraged. Client involvement in gover-
nance of service agencies will help to insure that clients contribute to the
way street-level bureaucrats define their roles. Service provision should be
decentralized to a significant extent, so that the advantages of orienting prac-
tice toward local initiatives can be realized.
During the 1g6os some communities experimented with client partici-
pation in governance of schools, neighborhood health centers, public hous-
ing, and other public services. From these experiences we should know bet-
ter than to encourage citizen control without examining the conditions of
transfer and the degree of control. More often than not, the experiments of
the 1g6os inappropriately discredited citizen participation by providing con-
trol over programs lacking financial viability or by narrowly circumscribing
the scope or powers of client or citizen boards. While avoiding the problems
of cooptation of community activists in financially unhealthy public en-
terprises, or of tokenistic participation, client control over service bureau-
cracies remains potentially critical in making the bureaucracies more respon-
sive to clients.
Issues of client control over service bureaucracies are not separable from
considerations of large-scale social changes or changes in the organization of
public services. It is surely simpler to contemplate neighborhood represen-
tative councils governing outreach health centers, legal services programs,
and local schools than it is to consider citizen participation in the governance
of centralized health facilities, downtown legal services complexes, and con-
solidated schools. Yet, if client control represents a way of making clients
more central to street-level bureaucrats' tasks and promoting further
changes by fundamentally challenging the bases of service provision, then
extending citizen control over complex systems and facilities should be
encouraged.
196
Support for Human Services
ble to circumscribe it. It is hardly obvious that every discretionary role
played by street-level bureaucrats should continue to exist. Where workers'
discretion leads to unfair and unequal treatment of clients, with no compen-
sating benefits, it should be desirable to reform systems by removing this
unredeemed source of unfairness. If, for example, a teacher's right to strike
pupils is judged to be undesirable in all cases, it is obviously appropriate to
prohibit flatly this ?Ption.
Some situations may arise, or may develop from previous practices, in
which the judgment must be made that intervention by street-level bureau-
crats is harmful or wasteful. However, the judgment that street-level bu-
reaucrats' discretion is inappropriate is not necessarily easily made.
The decision to separate social service functions from income-support de-
termination functions in public welfare is an example of removing consider-
able discretion from street-level bureaucrats. The stated intention was to
limit discretion of case workers in support determination and free them to
provide services without the burden of client appeals or tactics designed to
obtain more benefits. (Perhaps the primary appeal of this "reform" was that
it also promised to reduce the number of welfare personnel.) This change in
policy probably deserves support because the capacity of social service work-
ers to provide meaningful assistance to clients had become so circumscribed
that their interventions had largely lost whatever beneficial potential may
have existed in the previous definition of social worker roles. Moreover, the
paternalistic and degrading assumptions of the welfare system virtually
guaranteed that social services would be provided in a nonreciprocal con-
text. In this case the discretionary opportunities of case workers had already
become so limited by the social context in which aid was provided that it was
probably beneficial to further routinize the aid process. 2
Still, some doubts linger. Questions remain concerning the meaning of
this "reform." To what extent was this policy adopted in order to reduce the
power of social workers, diminishing their role and also their numbers? To
what extent was the transformation of support determination motivated by a
desire to reduce support levels by eliminating client-oriented social workers
from the process in favor of cost-conscious accountants? To what extent did
this reform indeed eliminate discretion, or did it simply transfer discre-
tionary powers to a new set of employees? Finally, by eliminating the social
service provider at intake, to what extent did this reform represent an im-
plicit decision to reduce service levels by restricting social workers' opportu-
nities to pick up relevant problems at intake, when people are often most
receptive to assistance?
Some of these questions arise generally in considering systems to rou-
197
THE FUTURE
199
THE FUTURE
200
Support for Human Services
lem solving, then training oriented toward transforming jobs into something
else is likely to be ineffective. 6 Thus one would be skeptical about develop-
ing new kinds of street-level bureaucrats primarily through instructional
efforts.
These reform orientations have their limits. Most street-level work is not
open to meaningful revision by limiting discretion, removing public em-
ployees from interaction with clients, or modestly altering bureaucratic
structure. In street-level bureaucracies there is an irreducible requirement
that public employees interact with citizens to determine the nature and ex-
tent of public services they should receive and to provide those services
through interactions with them. If this is the case we are challenged to ob-
tain accountability when bureaucratic mechanisms are inadequate and
inappropriate.
The limits on bureaucracy in this area draw attention to the argument that
street-level bureaucrats should be professionals whose relatively altruistic
behavior, high standards, and self-monitoring substitute for what the society
cannot dictate. Who will watch the watchmen? The watchmen will watch
themselves. The argument for professionalization comes down simply to the
realization that control of occupational groups must come from within the in-
dividual members of the group if it cannot be dictated from outside. If street-
level bureaucrats cannot be restricted in everyday functioning, then self-
monitoring must substitute for bureaucratic controls. To this extent, the
advocates of greater professionalism in street-level bureaucracies appear to
have an unassailable point. There is a powerful coincidence in our apparent
need to have street-level bureaucrats monitor their own performance and in
the aggrandizing claims of street-level bureaucrats that they are indeed suf-
ficiently autonomous and self-policing to be granted at least a degree of pro-
fessional status. 7
The professionalization of street-level bureaucracies is commended by
some analysts because standardized formal training in universities, seeking
training to get credentials, and control over occupational entry is already far
advanced in teaching, nursing, social work, and other street-level occupa-
tions. 8 In addition, it appears possible to influence the professionalization of
these occupations. Public policy to influence the direction of profes-
201
THE FUTURE
202
Support for Human Services
away from a service ideal or abandon it after protracted struggle with them-
selves or others, the careers of idealistic professional recruits are usually
abandoned to processes that insure their socialization to the dominant pro-
fessional values. There are at least three areas in which the professions as
currently organized help to erode the service orientation.
First, professionals by definition are accountable only to peers. While the
peer orientation protects professionals from the criticism of untrained and
inexperienced outsiders it also insulates them from the criticism of clients
and people who would speak on clients' behalf. Professionals are notoriously
reluctant to criticize each other and at best direct attention only to the most
extreme violations of ethical norms. Informal peer review is normally
avoided and formal peer review focuses on immoralities unrelated to profes-
sional performance or to narrowly defined technical capabilities. Formal
peer review is irrelevant to guiding the normal routines of practice or the di-
rections of the professions.1o
Of greater impact overall are informal peer pressures (as opposed to re-
views) that guide professional development. Here, as in other social groups,
professional novices are initiated into the rituals of practice through encour-
agement to develop amiable relations with peers, to protect fellow members
of the society, and generally to appreciate the problems of other profes-
sionals, even at the possible expense of clients. Thus the doctor's advocacy of
patients' medical needs is tempered by the requirement that he or she ap-
preciate the organizational needs of the hospital. Thus the lawyer's advocacy
of clients' rights is tempered by his or her need to appreciate the norms of
the court. Untempered peer definition of professional norms thus effectively
erodes the client orientation to which professionals are theoretically
committed.
Second, a major problem with professionalism as a model for street-level
bureaucracy is the tendency of professionals to work in isolation. Effective
professional norms call for mutual deference and shows of respect except
when other professionals grossly violate these norms or threaten to embar-
rass the profession. It is not that well-qualified professionals actually respect
the bumbling fool in their midst. However, it is unlikely that they will take
steps to help the fool improve (or, for that matter, to aid the eager but inex-
perienced novice) except when pressed by matters of self-interest. Looking
at the same problem from the viewpoint of the professional who could use
help, the norms usually inhibit professionals from seeking guidance in solv-
ing problems or providing services to clients, since to ask for help would be
to admit a degree of incapacity. It is instructive that police officers who feel
themselves dependent on their partners for personal safety are extremely in-
203
THE FUTURE
During the 1g6os the service occupations and professions appeared to expe-
rience a degree of revitalization as an influx of idealistic recruits joined with
like-minded members of the profession to dedicate themselves to practices
based on different principles from those that were dominant at the time.
These principles included placing their skills in the service of those people
who most needed them (the poor, ethnic, and racial minorities), committing
themselves to respecting client autonomy, turning the power of the profes-
sions to helping achieve greater social and economic justice, and eschewing
personal status enhancement when it conflicted with these principles. These
"new" professionals were radical in the classic sense of seeking a return to
first principles. 11
The surge of idealism during this period differed from other times only in
volume and attention paid by others. In every era people sympathetic to
these principles have entered the professions only to be ground down by the
social structure of their jobs. If there was a difference in kind in the 1g6os it
stemmed from the greater numbers, which provided a critical mass to articu-
late these ideals, and from the social movements that provided the setting in
which these ideals were stimulated and received notice and approval.
In every era, there is a propensity among at least some members of street-
level bureaucracies to work according to the ideal standards of their roles. If
204
Support for Human Services
we must depend upon a core of street-level workers who will strive to main-
tain integrity in the exercise of discretion, we may well ask what can be done
to support and enlarge this core? Where such a core does not presently exist,
we may ask what can be done to bring it into being? What can be done to
keep the new street-level bureaucrats flexible in their response to clients
and zealous in their commitment to client rights while delivering public pol-
icy? While pursuing these objectives, what can be done to insure that the
new street-level bureaucrat possesses the skills to intervene with clients ef-
fectively? It is helpful to ask these questions in these ways because they
direct attention to building on opportunities that currently exist. They guide
analysis of one possible direction of street-level reform without directly con-
fronting the enormity of attempting to think through fundamental refor~
for the totality of street-level bureaucracies.
The new street-level bureaucrats already exist in public agencies, trying as
best they can to maintain high standards and resist the routinization that
they fear is their fate. The new street-level bureacrats already exist in profes-
sional schools, searching for positions that will permit them to develop ca-
reers consistent with the objectives that motivated them to consider public
service in the first place. The new streevlevel bureaucrats exist in the reser-
voir of young people who would commit themselves to public service if they
had effective service models which they might follow.
What is needed to develop this potential?
Financial support for the human side of street-level bureaucracies is a nec-
essary although insufficient condition. Although additional resources cannot
overcome the patterns of routinization and simplification that are currently
endemic, teachers, legal services lawyers, police officers, and other street-
level bureaucrats will not have the slack to organize themselves for more
responsive interactions with clients unless they have adequate material sup-
port. It is particularly important to reverse the current decline in support for
the human services, since workers' impressions of harassment and resource
inadequacy are probably as important as the fact when it comes time to orga-
nize client processing. Public agencies must provide an atmosphere of delib-
eration or workers will not be able to escape the conviction, rooted in coping
needs, that they must routinize client processing. This is part of the reason
that incremental reductions in case load often fail to-show an effect, since the
feeling of harassment remains when case loads are marginally reduced, say,
from so to 45 cases.
Financial resources are also necessary to provide the incentives necessary
to make possible career commitments. It is not that a new street-level bu-
reaucracy needs to provide the same financial rewards as careers in the
205
THE FUTURE
206
Support for Human Services
takes place in the schooling of rookies in how to survive on the street, with
supervisors, or in the face of client harassment. In other words, peer review
and instruction currently do take place, but in ways that force workers either
to be extremely circumspect or to promote routine processing rather than
responses appropriate to individual clients.
The development of peer support mechanisms can and must be related to
work processes. Street-level bureaucrats need to receive recognition for
good work and to be free to seek help when they encounter work-related dif-
ficulties, without feeling that their reputations are in jeopardy. Perhaps out-
side specialists should systematically review the work of street-level bureau-
crats with their clients. Perhaps agencies can develop this capacity without
such assistance. Whatever the mechanism, those street-level bureaucrats
who continue to aspire to provide appropriate community service will wel-
come the chance to grow in their jobs without being judged and placed at
risk in the process.
A street-level bureaucracy that has developed processes of staff growth
and development will also develop processes for small group decision mak-
ing. Small group units for street-level decision making (for example, grade
levels or departments in schools, subprecinct units in police departments,
neighborhood service offices in legal assistance and health maintenance) are
probably best suited to determine which aspects of social services should be
routinized and which aspects should remain unprogrammed. Routinization
in social life may be inevitable, but it is not inevitable that routines should be
imposed from above or by authorities that do not directly confront clients.
Decentralized units would be far more likely to develop routines consistent
with responsive and efficient client treatment than authorities removed from
the scene, particularly if outside audits are continually able to draw attention
to issues of service quality.
Fundamentally at issue is making the most of the reality that street-level
bureaucrats primarily determine policy implementation, not their superiors.
If the bureaucratic connection between lower-level workers and the reins of
authority are indeed tenuous, as I have argued, perhaps it is better to Aow
with the organizational dynamics of policy delivery in these organizations
rather than resist them by insisting on bureaucratic solutions to problems
defined as worker deviation from preferred performance.
In reality, decentralized units given full responsibility for practice would
have to resist the tendency to drift toward recentralization of routine func-
tioning. The pull would be strong to let higher authorities make critical
decisions, thereby absolving lower-level workers of responsibility. How-
207
THE FUTURE
ever, even creating the opportunity for self-determination of small units pro-
vides a context for considerable learning and the potential for achieving a
more client-oriented practice.
Even if it were possible to restructure street-level bureaucracies in these
ways, what would keep people from drifting into the old patterns? This is a
particularly significant question if we believe that bureaucracy to some degree
reflects as well as reinforces and perpetuates the prevailing social structures.
Three considerations conducive to workers' exercising effective and respon-
sible control over the work situation m,ay help consolidate changes in street-
level bureaucracies in ways that limit the likelihood of retrenchment.
First, the clients of service must become a more potent force in the refer-
ence groups of street-level bureaucrats. Ways must be discovered to make
visible and accessible the behavior of lower-level workers, and clients likely
to be affected by their actions must become more involved in the definition
of good practice. To the extent that peer relations are the primary source of
the expectations and values promoted within the occupational sector, there
will remain a temptation to develop esoteric criteria of practice judgment.
Street-level bureaucrats' performance has not been so good nor our con-
fidence in their work so well established that great harm would come from
creating regular mechanisms to expose street-level work to the scrutiny of
clients. Even if clients are overwhelmed by the trappings and rhetoric of
professionalism or are limited in their understanding of the ramifications of
decision making, exposing the decision-making environment to clients
should anchor street-level bureaucracies more firmly in a client orientation.
Moreover, street-level bureaucrats should undertake to develop techniques
to educate clients toward making better judgments about seeking service
and better assessments of service provision. Studies and observations con-
cluding that clients are overwhelmed by professionals caution us about in-
volving clients in decision making, but they do not reflect experiences in
which client involvement has been systematically nurtured. 12
Client contributions would be enhanced if street-level units accepted re-
sponsibility for group case loads rather than incorporating clients as the case
loads of individual workers. In many street-level bureaucracies, a primary
contribution to workers' isolation and pressure is the fact that workers are in-
dividually positioned to be fully responsible for clients, are unable to seek as-
sistance or advice, and must compete with other workers for advantage so as
to minimize their load. So long as street-level workers are individually re-
sponsible for their sector of client services they are likely to be defensive in
developing cooperative and supportive relations with fellow workers or
clients. Without abandoning the efficiencies of specialization or the account-
Support for Human Services
ability that individual case loads minimally provide, it is possible to develop
conceptions of group or office case loads that make clients the responsibility
of the staff, not individual workers. 13
A second requirement in sustaining the new street-level bureaucracy is
the zeal and leadership of people committed to the new orientation. Reform
orientations are not self-implementing. They can only survive in a context in
which people are dedicated to public service and receive support from client
groups, fellow workers, and the community. The nurturing of such leaders is
a process that would well begin in the training grounds of universities,
where relatively visionary orientations are sometimes rewarded, and would
continue through a public policy that valued such leaders for their commit-
ment to a client-oriented service. Without the development of such a eward
structure it would be reasonable to conclude that there was no constituency
for these reforms.
Traditionally universities have provided strategic sanctuary for some of
the most important dissenters from contemporary practice, but they have
often been rendered ineffective because their lessons come from the ivy
tower rather than from the streets. This is particularly regretable because
such teachers may inspire young professionals to go into their work without
expe!:iencing the dilemmas of practice or helping to prepare the environ-
ment into which students insert themselves. Ihe new street-level bureau-
cracies would be significantly assisted by policies in which supporters of this
orientation circulate between the teaching of young professionals and the
practice of public service. Some of their teaching ought to be done not in
universities but in the field, where there is opportunity for constant confron-
tation with the realities of practice. A police academy in which students
worked half-time as apprentice officers, or a social work institute in which
students received training from teachers who shared an office practice with
their students, might provide the reality-based environment in which staff
and students would find an appropriate balance between experience and
detachment.
A final aspect of support for maintaining the client orientation of street-
level bureaucracy rests in the development of ongoing processes of suppor-
tive criticism and inquiry. Built into every week of practice should be oppor-
tunities to review individuals' work, share criticisms, and seek a collective
capacity to improve performance. The orientation should be skeptical, for
the objective of such sessions would be to resist where appropriate the early
closure of possibilities that accompanies the inevitable routinization of prac-
tice. Some police departments hold an inquiry every time an officer draws a
weapon. Hospitals ordinarily inquire into circumstances accompanying
209
THE FUTURE
210
Support for Human Services
only when an effective coalition develops that harnesses public concerns for
service costs and effectiveness, respects client involvement in service proce-
dures, and recognizes the needs of the work place, where the fate of innova-
tion will ultimately be decided. This is not likely to occur in this society of
protected interests unless this social and political movement brings the pri-
ority of more humane service provision to the forefront of concern. If such
concern is likely to be evident in the future it will be because place by place
and issue by issue, people-effectively demand respect for themselves and
their proper claims on government, while at the same time they are able to
explore ways to support street-level bureaucracies in their struggle to do a
decent job under adverse circumstances.
211
CHAPTER 14
On Managing Street-Level
Bureaucracy
212
On Managin g Street-L evel Bureauc racy
written as African Americans and their allies sought to bring into reality the
promises of equal rights. In the early 1970s, when the structure of the book
was being worked out, the ripples of the civil rights movement included con-
flicts over schools, the police, and paternalistic welfare administration. These
institutions were the places people encounter ed or confronted government,
the places unequal treatment would be manifest. In general, they were pop-
ulated by white workers and administrators who became the focus of conten-
tious race relations.
Since the early 1g6os I had been personally and professionally committed
to advancing the objectives of the civil rights movement, as I understood
them. But this did not include sympathy for characterizations of public sec-
tor workers as indifferent to the experiences of the people they dealt with. A
crude racism could not explain on-the-job behavior. Nor did the many inter-
views and observations of street-level bureaucrat s my collaborators and I
conducted support the proposition that personal attitudes or indifference of
street_-level workers could explain the experiences of citizens with these
agencies.
I wanted to understand the performan ce of critical public services that
were either at the heart of community conflict or were central to providing
for citizens in need-the panoply of institutions that focus on citizen wel-
fare: income supports, housing provision, social services, and the justice sys-
tem. I came to believe that the experiences of citizens with public services
were the predictabl e consequen ces of the structure of particular kinds of
publics sector work. This belief was based on the recognition that the work
of police officers and then, I realized, other frontline workers, consisted of
layers of coping mechanisms that guided them through complex interactions
with the people they served, and allowed them to salvage at least a portion of
their work as authentic and worthwhile. My primary concern became recon-
ciling the human dimension of street-level positions with the often problem-
atic outcomes of the systems in which they worked.
Another influence was the emergence in political science of studies that
analyzed what actually happens when public policies are enacted-w ork that
later became known as implement ation studies. It was only in the mid-1g6os
that political scientists began to focus on the now-obvious point that the pas-
sage of laws and the promulgation of regulations did not fully encompass the
policy-making process. On the contrary, as we now take for granted, policy is
only finally made when laws or regulations are fully implement ed, through
subsequen t processes that cascade from the initial declarations.
At least initially, studies of implement ation were part of a discourse that
sought to confront the contradictions between American ideals and Ameri-
213
THE FUTURE
can policy in practice. In 1968, as the issues contested in the streets began to
engage academics, James Davis and Kenneth Dolbeare published a study of
the Selective Service System that asked whether the military draft fairly ap-
portioned risk among potential recruits, as it was supposed to do. 1 They
found that the implementation of the military draft laws led to systematic
overrepresentation of young men from lower socioeconomic backgrounds,
despite elaborate propaganda to the contrary. Studies followed that sought to
understand why substantial plans to enhance economic development came
to grief in Washington, D.C., 2 and Oakland, Califomia. 3 These early studies
were products of a broad skepticism about the capacity of government
shared by many elements in society in the decade deeply influenced by the
"discovery of poverty" amidst plenty, the civil rights movement, ghetto upris-
ings, and the antiwar movement.
Imbedded in the early works on implementation is an implicit critique of
American democracy and an appeal to close the gap between public prom-
ises and performance. Thirty years later, concerns about access to public ser-
vices, and of the relationship between public promises and policy implemen-
tation, continue to be valid. But they are joined by another compelling issue.
We now live in an age in which support for government has been eroded,
and the very purpose of government is deeply contested. The times require
advancing an understanding of the critical roles of government on which
successful modem societies depend: social welfare provision, environmental
and consumer protection, public goods such as schools and highways, and
planning and investing for the future. Appreciating the complexity of public
services through which citizens engage government on a regular basis should
make a significant contribution to this agenda.
The new reality results from several trends that have been well-rehearsed
elsewhere but bear mentioning here. For one thing, today we confront limits
that were hardly on the horizon thirty years ago. In the post-World War II
period, an era of "easy financing" facilitated the expansion of public pro-
grams with only limited concern for whether revenues could be found to
support them. 4 That era is long over. In state governments that by law can-
not run deficits, public budgeting has always been about playing zero-sum
games. Now this is the case at the federal level as well, as federal officials are
under increasing pressure to match any new expenditure with a correspond-
ing budget cut or tax increase.
Another major development has been the sharpening of the tools of gov-
ernment and the invention of new tools. In contrast to earlier times, when
the distinction between governmental and nongovernmental was reasonably
clear, public policies today are coproduced with individuals through tax
214
On Managi ng Street-L evel Bureau cracy
breaks, credits and vouchers; and with private organizations through tax ex-
penditure s, subsidies, public-private partnersh ips, and contracti ng for ser-
vices.
Separate from these developm ents but related to them are two additional
trends that shape the future of street-lev el bureaucracy. The rise and persis-
tent strength of the conservative perspecti ve on public policy, through aca-
demic and scholarly outlets as well as the pronounc ements of political lead-
ers, has promoted the view that governm ent is intrinsically inefficient and
inherentl y wasteful. According to this perspecti ve, governm ent should be
small and taxes low. In addition, markets and market mechanisms, in con-
trast to governm ent, are said to produce optimal outcomes and lead to inno-
vation and lower costs.
In hindsight I see that in the earlier edition of Street-Level Bureaucracy I
took the existence of governm ent and critical public services for granted, as
most comment ators still do today. I assumed that governm ent services,
though they might change to some degree, were fundamen tally enduring in
their basic structure and support. With the passage of time those assump-
tions seem more tenuous, for several reasons:
1. Contracting for services and other new approaches to service delivery,
which app~ar to introduce market-like mechanisms into public systems,
have been hastened onto the public stage by the conservative challenges to
existing forms of service delivery.
z. At the national and particularly the state and local level, where education,
policing, income supports, and other social policies are administered, gov-
ernments are more or less in permanent fiscal crises as they constantly
struggle to match revenues with the cost of meeting existing obligations.
5
215
THE FUTURE
FISCAL STRINGENCY
Because state governments cannot run deficits and their tax revenues vary
with employment, wages, and consumer spending, they are subject to budget
fluctuations as revenues rise and fall with the business cycle. Consequently,
they periodically need to cut budgets as revenues decline in economic down-
216
On Managing Street-Level Bureaucracy
turns. Over the years, as the anti-tax, small-government perspective has taken
hold, state governments have been thrown into more or less permanent bud-
get crises. They have legislated spending limits and passed laws making it dif-
ficult to raise taxes, or have just cut taxes when there have been surpluses,
and cut programs when revenues declined. 8
It has been in the context of continual pressure on limiting the growth of
their budgets in relation to need that the states began to experience the
Great Recession in 2008. One year into the recession, states overall have had
to cut almost one out of every four dollars in their budgets, with more cuts
anticipated for the following year. 9 Such cuts take enormous tolls on the ser-
vices that provide for or protect people in need through street-level bureau-
cracies.
In Ohio, to take one set of examples, a series of tax reductions and a re-
fusal to raise taxes in a recession led in 2009 to the elimination of the state's
early childhood education initiative, and to deep cuts in subsidized child-
care, mental health services, and protective services for children and adults. 10
In October 2oog, the eminent economist Paul Krugman calculated that over
the previous five months the United States had fired more than 143,000
teachers in reaction to the recession, with undoubtedly tens of thousands
more on the block as state revenues continue to be in crisis. 11
No politician wants to reduce the number of teachers, and few want to
reduce the number of other public employees-recreati on workers, nursing
home attendants, and the like. These people work in services of which the
public generally approves. 12 Instead, politicians target the size of govern-
ment overall, making cuts in programs the inevitable consequence of re-
duced revenues. Writing about public benefits at the national level, political
scientist Paul Pierson observes this dynamic: "There is a fundamental asym-
metry between the organized advocates of public spending, who favor par-
ticular governmental initiatives, and their opponents, who (at least rhetori-
cally) criticize government spending in general and on principle." 13
CRITICISM OF GOVERNMENT
Casting a long shadow over the provision of public services is the rise and
influence of the conservative movement, supported by well-funded think
tanks and university centers. Since the 1970s and accelerating in influence
during the Reagan era, conservative theorists and political figures have re-
lentlessly promoted a set of linked ideas to support their version of a better
society: small government, low taxes, reliance on individual initiative (except
for subsidies and tax breaks for favored interests), and reliance on market
mechanisms to achieve efficiency and innovation. Over these years the con-
217
THE FUTURE
218
On Managing Street-Level Bureaucracy
posing them in practice, and news about them focusing only on their failings,
it is hardly a wonder that the reputation of government is low.
Of course, even democratic governments, like all human endeavors, are
sometimes flawed, just as some churches are run by people of questionable
morals, and some corporate leaders bring their companies to bankruptcy or
breach the public trust. But of the major institutions in society, only govern-
ment sustains attacks without effective rebuttal or even measured assess-
ment of the charges.
The good news, I suppose, is that public programs typically enjoy popular
support and are relatively stable over time. New programs are introduced as
the range of problems deemed to require public action expands, and as new
populations are included in the broad consensus of who deserves assistance.
Although recent decades have witnessed vigorous efforts to cut back the
state, the basic composition of the broad social compact remains intact. 14
Unfortunately, although members of the public consistently affirm their sup-
port of the programs government sponsors, their initial reaction to govern-
ment in general indicates broad acceptance of the antigovernment perspec-
tive. A striking illustration of this was revealed in a 1997 Kaiser/Harvard poll.
When asked their opinion about a health plan hotline, 70 percent were
strongly in favor. The number dropped precipitously, to 43 percent, when
the question asked about a government hotline. 15
The shift in the political discourse opens several new paths to what, how,
and whether government will respond to social problems and closes off oth-
ers. The strong antigovernment perspective throws advocates for public ser-
vices constantly on the defensive. Drawing the skirmish line in a public de-
bate at whether something is "governmental" at the least ensures that policy
debates will be ideological (what are the universal principles and values in
terms of which the policy should be assessed?) rather than pragmatic (what
are the best ways to solve this particular problem?).
One reason public managers have been put on the defensive is recent cri-
tiques of government effectiveness that draw on a narrow reading of busi-
ness practices. Demands for accountability often seem to be attacking gov-
ernment programs even as they profess interest in improving them. To
paraphrase Murray Edelman, the insightful guide to politics as symbolic ac-
tion, to insist that one needs greater accountability is also to caution that ac-
countability is lacking. 16
A case in point is the movement to demand that social programs be sub-
ject to the highest evaluation standards, of which controlled experiments
with randomized assignment of subjects is the gold standard. Every policy
analyst recognizes that such research, when appropriate and properly exe-
219
THE FUTURE
220
On Managing Street-Level Bureaucracy
are not just narrow matters of achieving more effective public services at the
appropriate cost. They may also be understood as contributing to a more
substantial agenda in which government, by improving its public services
across all the divides of race, ethnicity, and class, is perceived as fair and
trustworthy.
In the next section I treat the central question of whether and how street-
level bureaucrats can be managed when their work is defined as exercising
discretion. I then take up the challenge of improving the capacity of street-
level bureaucrats to make sound judgments when relatively standard ap-
proaches to management are insufficient in achieving public purposes.
221
THE FUTURE
any policy reform, we need to look into the entire policy environment in
which street-level bureaucrats function.
The transformation in 1996 of the basic American welfare program was
obviously a critical policy decision that took place far beyond the interview
rooms of local welfare offices. The new program ended welfare as an entitle-
ment and made eligibility depend on work effort and job-seeking. Street-
level bureaucrats, who only a few weeks earlier were focusing on monitoring
the accuracy of applications and applying eligibility rules, were now required
to learn the vocabulary and ways of a system that expected clients to look for
work and find it. And, ideally, they would have to learn how to apply that
vocabulary to welfare applicants as they presented themselves, and strategize
about the most appropriate path for clients to enter the workforce. These
changes reflected policy making at higher levels. In addition, options avail-
able to workers in processing their caseloads were heavily influenced by sev-
eral factors.
These (and other) factors shape the decision making of today's welfare
workers. As Evelyn Brodkin observes, "the quality of choices about help
[available to workers] depend[s] in part upon the 'helping resources' to
which street-level bureaucrats have access." 22 As always, they depend on
caseloads and other work circumstances that impinge on conditions of work
in the welfare system. But since welfare reform they also depend on some
new factors-especially the local employment situation-that are beyond
their control.
Consider the New York Times report in 2009 that "more people were be-
ing turned down for welfare benefits because of overly stringent or confus-
ing requirements."23 It may be that welfare workers were exercising exces-
sive discretion in discharging their responsibilities, but it also could be that
the rules in New York really were overly stringent or confusing.
In short, in the new work and welfare systems, workers categorize appli-
cants and deploy available resources according to routines and simplifica-
tions they develop, even as they exercise judgment and make unsanctioned
222
On Managing Street-Level Bureaucracy
responses. The content of these discretionary acts will not be random, and
they will not only be related to the coping needs of the workers. On the con-
trary, they will be structured by the choices available to the workers as op-
tions (such as local funding for daycare) provided by the policies.
To understand worker behavior we need to pay much more attention to
the overall structure of the policies. A study of nurses' preferences for public
versus private hospitals in Australia concluded that the shortage of nurses in
public hospitals might be attributed to their having less autonomy than their
counterparts in private hospitals. The hypothesis is plausible, but other im-
portant factors, which the study did not investigate, might have accounted
for the nurses' job preferences, including whether the private hospital nurses
were paid higher wages, worked better or fewer hours, served a more conge-
nial clientele, or had lower caseloads.
223
THE FUTURE
224
On Managing Street-Level Bureaucracy
cannot use them, and that language barriers and other issues of citizen ac-
cess are addressed.
225
THE FUTURE
ET established a new mission for the welfare department and recruited its
workers to support the mission by relentlessly promoting indicators of sue-
227
THE FUTUR E
cess, keeping track of performance , and conveying to workers that they were
critical to the new mission of the agency. 37
CompSTAT. A successful and widely imitated initiative that appears to
embody these principles is CompSTAT, the police administrative reform be-
gun during the administration of Commission er William Bratton in New
York City in 1994. 38 CompSTAT takes advantage of high quality data collec-
tion and reporting of crime statistics at the precinct level to hold precinct
commanders accountable. Precinct commanders , like the welfare adminis-
trators subject to quality control initiatives mentioned earlier, in tum engage
police officers to achieve the desired results.
In monthly meetings with the Police Department' s high command, ad-
ministrators require all precinct commanders to be ready to defend their
performance and explain trends in their districts that require attention. If
there has been a spike in reported burglaries, for example, the commander
can expect to be interrogated on why this trend is occurring, and what he or
she intends to do about it. The system works when precinct commanders ,
with clarity about their priorities based on the precinct data, and urgency
deriving from the monthly meetings, design strategies to improve the perfor-
mance of their units. At a certain level the critical elements present in the
ET case are present here as well: clarity of mission, close tracking of perfor-
mance, and keen involvement of the organization's leadership.
In evaluating these reforms a key question is whether focusing on a spe-
cific aspect of the mission pulls street-level bureaucrats away from the full
range of their responsibilities to a detrimental degree. Does monitoring and
evaluating performance on a particular measure improve job performance
overall? Or does focusing on a single dimension of the job subtly change the
job itself in negative ways?
In CompSTAT, police are not relieved of responsibility for law enforce-
ment outside of departmenta l priorities, and in any case, if new crime trends
emerge, they also can be targeted. In the ET experience, job placement was
only one of many departmenta l priorities that received attention through
monitoring and feedback from the high command. For example, in ET
maintaining a good record on quality control remained a specified agency
priority even though the first priority was recipient employment . CompSTAT
has been widely emulated in other police departments , and its principles
have been widely adopted in efforts to improve accountability in other agen-
cies and-in the cases of Baltimore and Washington State-gover nment-
wide initiatives. As with any replications, success in these efforts depends on
how faithfully the model is actually reproduced. 39
These reforms of street-level bureaucracy seek to harness measuremen t
On Managing Street-Level Bureaucracy
of performance with appropriate sanctions and rewards. Their success ap-
pears to depend upon the alignment of the new approach with the core val-
ues of the agency, the clarity of the mission, and the extent to which street-
level bureaucrats can in fact improve their performance in pursuit of the
mission. 40 These examples demonstrate that managers can influence the be-
havior of street-level bureaucrats by strategically choosing the data they col-
lect about workers' performance, and the incentives with which they respond
to the data.
229
THE FUTURE
people in need despite rules to the contrary, they join what sociologist Lisa
Dodson smartly calls "the moral underground."42 Members of the moral un-
derground do not blow up rail lines. What they do is ignore rules that they
are supposed to enforce but seem excessively rigid, in order to help people
in need, at little overall cost to society.
Acts of flexibility not only benefit the recipient of compassion but also
have larger implications for society. After presenting a case in which a voca-
tional rehabilitation counselor cut through red tape to obtain a computer for
a client, Steven Maynard-Moody and Michael Musheno describe the coun-
selor as engaging "in an act that redeems the state by breaking through the
bureaucratic labyrinth." 43 In addition, they suggest, such an encounter pro-
vides a context in which community values that coexist and sometimes con-
flict with bureaucratic rules find expression. Many administrative systems al-
low for such flexibility under certain circumstances, though flexibility can be
misused and can result in wholesale undercutting of intended policy. Like so
many matters in policy analysis, the proper degree of flexibility is a matter of
balance.
Treating everyone alike is a requisite of building popular trust. But good
rules allow for appropriate exceptions. How, then, can a society tolerate be-
nign exceptions when universalism is so important? Exceptions that favor
people because of their social ties or ethnic backgrounds undermine popular
trust in government's fairness. We cannot specify where the lines should be
drawn, but we can acknowledge that responsiveness within a norm of univer-
salistic rule application can be consistent with fair and effective public ser-
vices, even if it requires departure from strict mechanical accountability. The
exceptions that mark the superior public agency must be those that would
pass a test of deserving popular approval on the basis of the presenting facts
and local values, and are not tainted by implications of favoritism based on
the identities of those favored.
A second circumstance in which public service bureaucracies must em-
phasize responsiveness is when public services requires individual initiative,
the cultivation of experience, and a degree of empathy that cannot be re-
duced to administrative guidelines. Thus there is another '"problem" of
street-level bureaucracy that is entirely different than the one we have con-
sidered up to now: how to insure that people employed by the state to teach,
judge, evaluate, and counsel have the necessary skills, experience, and train-
ing to exercise discretion properly and most effectively. For some street-
level occupations more than others, it is essential that decisions are made
fully in response to the individual case or situation, deploying deep knowl-
edge of the field with commitment and a common sense that cannot be codi-
230
On Managing Street-Level Bureaucracy
fied. As Michael Hill and Peter Hupe put it, "[e]nhancing street-level discre-
tion may, under certain conditions, be more functional for the implementation
of those policies than curbing it." 44
231
THE FUTURE
the tests in order to ensure that more schools meet the minimum stan-
dards.48
A deeper problem with NCLB, and one that generally confronts perfor-
mance measurement reforms, is the failure to address the strategic question
of how to improve achievement. In principle, performance-based account-
ability leaves open the question of how results are achieved; encouraging
flexibility in achieving goals is more or less the point of these approaches.
However, if the construction of a performance-based regime absorbs exces-
sive resources and the attention of administrators, creating ways to achieve
better results may suffer. This point is particularly relevant to NCLB, given
the substantial evidence that improving the capacity and training of teachers
is one of the most powerful paths to improved educational outcomes. 49
NCLB requires the states to employ "highly qualified teachers," but to do
so schools need only hire educators in core academic areas who are licensed,
hold a bachelor's degree, and have taken courses in their subject area (as evi-
denced, say, by their undergraduate major). This is a decidedly shallow stan-
dard of what constitutes "high qualification" of a good educator. Higher ex-
pectations of teacher training and preparation would yield much better
results. Recommending the approach of the National Board for Professional
Teaching Standards (NBPTS), Mary E. Dilworth and Joseph A. Aguerrebere
suggest that highly qualified teachers must meet the following criteria.
1. They are trained in their subject areas and broadly grounded in the liberal
arts so as to be able to help students make connections across issue areas.
2. They command a variety of approaches to learning and are able to adapt
their approaches to students with different learning needs and styles.
3 They are committed to teaching and the belief that all students can leam. 50
The last is not a small matter and is separable from subject area knowl-
edge and teaching methods. Good teaching depends on dedication and on
the teacher's conviction that students will be able to succeed. Students who
are taught by teachers certified by NBTPS, in 2009 numbering almost
83,000, achieve at a higher rate than those taught by uncertified teachers, as
measured by evaluations in North Carolina, Arizona, and Florida. 51
In principle, all street-level bureaucrats, from health-care workers to po-
lice officers, must be able to assess the presenting situation or client freshly.
Equal treatment may require treating people differently to achieve equal re-
sults, particularly when responding to diverse populations. In the case of
teachers, according to Dilworth and Aguerrebere, this means that teachers
should know their students well, both as individuals and as children from
shared but varying backgrounds. Teachers who apply "culturally responsive
232
On Managing Street-Level Bureaucracy
pedagogy" will understand, for example, "what types of errors a student who
is learning English is likely to make based on patterns that exist in their na-
tive language."52
Accountability systems with single indicators are magnets for criticism be-
cause public sector services rarely have unitary goals. 53 The challenge for
critics and supporters of No Child Left Behind is whether reducing the mis-
sion of the schools to achievement on the current tests is an acceptable re-
conceptualization of our expectations for schools. A generous assessment of
NCLB is that it has placed achievement of every student at the center of a
national debate, and located the problem at the level of the individual school.
As mentioned, the law is distressingly simplistic on what constitutes a good
education, and it is also more or less silent on how to achieve better results.
Despite substantial criticisms, the law will have proved valuable if future it-
erations lead to support for substantial investments in teacher salaries, work-
ing conditions, standards, and training.
233
THE FUTURE
234
On Managing Street-Level Bureaucracy
of being reprimanded. The consequences of making a mistake were so great
that workers who had committed "near-errors" never discussed them openly
or sought advice about how they might have handled situations differently.
DSS lacked the cultural norm of being able to ask for help or advice.
In response, Spence introduced a version of no-fault case review widely
adopted in U.S. hospitals after being pioneered in the Veteran's Administra-
tion.56 Under this model workers are encouraged to bring up troubling cases
for review. The model rejects the "error and blame" approach to account-
ability in favor of a culture in which it is safe to discuss and acknowledge er-
ror the better to encourage organizational learning. The no-fault model
works by making it safe to bring up problematic situations on the assump-
tions that no sanction will result from bringing up problems in the field, and
that such discussions are the basis for workers' and the organization's con-
tinuous improvement.
The second innovation was to introduce the team approach to case work.
Two workers rather than one would be responsible for each case. This ap-
proach would directly confront a critical imperfection in the existing system,
in which a single worker, with little experience and few resources to cope
with the tensions of the visit, would call on a distressed family. Family mem-
bers were often hostile or unresponsive in these settings, and often withheld
critical information about the functioning of the family.
A team approach might be better. No longer would an individual worker
be the solitary judge making a momentous decision alone. Workers could
share notes and consult with one another on a course of action. Families
were more forthcoming, perhaps because they believed they could appeal to
another worker if one did not seem sympathetic, or because the team ap-
proach conveyed to clients that the agency took their case more seriously. In
any event, the team approach found favor with workers and clients alike.
No outcomes in protective service interventions are certain, particularly
when the philosophy of child protection toggles between removing children
from the home and maintaining family integrity. Children die or are hurt
while families are under protective custody orders, just as they die or are
hurt in families not designated as requiring state intervention. In the new
approach to accountability, Spence promised never to scapegoat workers
when a child died in the care of the department-if the worker had made a
good-faith effort to execute agency policy. He acknowledged that the work
was fraught with danger and uncertainty, and said he would protect workers
when misfortune occurred, as it inevitably would.
This approach to reforming protective services recognizes that human
judgment is essential for effective public policy, and that the central chal-
235
THE FUTURE
Conclusion
237
NOTE S
Part I: Introduction
Chapter 1
1. These definitions are analytical. They focus not on nominal occupational roles but on the
characteristics of the particular work situations. Thus not every street-level bureaucrat works for
a street-level bureaucracy [for example, a relocation specialist (a type of street-level bureaucrat)
may work for an urban renewal agency whose employees are mostly planners, builders, and
other technicians]. Conversely, not all employees of street-level bureaucracies are street-level
bureaucrats (for example, file clerks in a welfare department or police officers on routine clerical
assignments).
The conception of street-level bureaucracy was originally proposed in a paper prepared for
the Annual Meeting of the American Political Science Association in 1g6g, "Toward a Theory of
Street-Level Bureaucracy." It was later revised and published in Willis Hawley and Michael
Lipsky, eds., Theoretical Perspectives on Urban Politics (Englewood Cliffs, N.J.: Prentice-Hall,
1977). pp. 1g6-213
2. U.S. Bureau of the Census, Public Employment in 1973, Series GE 73 No. 1 (Washing-
ton, D.C.: Government Printing Office, 1974), p. 9 Presented in Alan Baker and Barbara
Grouby, "Employmen t and Payrolls of State and Local Government s, By Function: October
1973," Municipal Year Book, 1975 (Washington, D.C.: International City Managers Associa-
tion, 1975), pp. 1og--112, table 4/3. Also, Marianne Stein Kah, "City Employmen t and Payrolls:
1975," Municipal Year Book, 1977 (Washington, D.C.: International City Managers Associa-
tion, 1977), pp. 173-179. These figures have been adjusted to represent full-time equivalents.
For purposes of assessing public commitment s to providing services, full-time equivalents are
more appropriate statistics than total employment figures, which count many part-time em-
ployees.
3 Jeffry H. Galper, The Politics of Social Services (Englewood Cliffs, N.J.: Prentice-Hall,
1975), P 56.
4 Lois Forer, Death of the Law (New York: McKay, 1975) p. 191.
5 New York Times, April4, 1976, p. 22.
6. Baker and Grouby, "Employmen t and Payrolls of State and Local Governments.
7 New York Times, July 10, 1977, p. F13.
8. Of four cities with populations over one million responding to a Municipal Year Book
survey, the proportion of personnel expenditures to total expenditures in police departments
averaged 94 percent and did not go below 86 percent. Cities with smaller populations showed
similar tendencies. These observations are derived from David Lewin, "Expenditur e, Compen-
sation, and Employment Data in Police, Fire, and Refuse Collection and Disposal Depart-
ments," Municipal Year Book, 1975 pp. 3!r98, table 1/21. However, the variation was much
greater in the less populous cities because of smaller base figures and the fact that when cities
with smaller bases make capital investments, the ratio of personnel to total expenditures
changes more precipitously.
That public expenditures for street-level bureaucracies go to individuals primarily as salaries
may also be demonstrate d in the case of education. For example, more than 73 percent of all
noncapital education expenditures inside Standard Metropolitan Statistical Areas goes toward
personal services (i.e., salaries). See Government Finances, Number 1, Finances of School Dis-
tricts, 1972 U.S. Census of Government (Bureau of the Census, Social and Economic Statistics
Administration, U.S Department of Commerce), table 4
239
Notes
9 Many analysts have discussed the increasing role of services in the economy. See Daniel
Bell, The Coming of the Post-Industrial Society: A Venture in Social Forecasting (New York:
Basic Books, 1973); Alan Gartner and Frank Reissman, The Service Society and the Consumer
Vanguard (New York: Harper & Row, 1974); Victor Fuchs, The Service Economy (New York:
Columbia University Press, 1968). On transformations in public welfare, see Gilbert Steiner,
Social Insecurity (Chicago: Rand McNally, 1g66), chap. 1; on public safety, see Allan Silver,
"The Demand for Order in Civil Society," in David Bordua, ed., The Police: Six Sociological
Essays (New York: John Wiley, 1g67), pp. 1-24.
10. Charles Reich, "The New Property, Yale Law journal, vol. 72 (April, 1g64): 733-787.
11. Carl Hosticka, "Legal Services Lawyers Encounter Clients: A Study in Street-Level
Bureaucracy" (Ph.D. diss., Massachusetts Institute of Technology, 1976), pp. 11-13.
12. See Frances Piven' s convincing essay in which she argues that social service workers
were the major beneficiaries of federal programs concerned with cities and poor people in the
1g6os. Piven, "The Urban Crisis: Who Got What and Why," in Richard Cloward and Frances
Piven, The Politics of Turmoil (New York: Vintage Books, 1972) pp. 314-351.
13. J. Joseph Loewen berg and Michael H. Moskow, eds., Collective Bargaining in Govern-
numt (Englewood Cliffs, N.J.: Prentice-Hall, 1972). A. Laurence Chickering, ed., Public Em-
ployee Unions (Lexington, Mass., Lexington Books, 1976); and Margaret Levi, Bureaucratic In-
surgency (Lexington, Mass.: Lexington Books, 1977).
14. The decline is a function of the lower birthrate and periodicity in the size of the school-
age population originally resulting from the birth explosion following World War II. See Baker
and Grouby, Municipal Year Book, 1975, pp. 109ff., on serviceability ratios.
15. This perspective remains applicable in the current period. However, in reaction to this
tendency, programs that would eliminate service mediators and service providers, such as nega-
tive income taxation and housing allowances, have gained support. Fiscal scarcity has brought to
public attention questions concerning the marginal utility of some of these service areas.
16. Consider the New York City policemen who, in October 1976, agreed to work overtime
without pay so that a crop of rookie patrolmen would not be eliminated. New York Times, Octo-
ber 24, 1976, p. 24.
17. There can be no better illustration of the strength of the organized service workers and
their support by relevant interests than the New York State Assembly's overriding of Gov.
Hugh Carey's veto of the so-called Stavisky bill. This legislation, written in a period of massive
concern lor cutting the New York City budget, required the city to spend no less on education
in the three years following the fiscal collapse than in the three years before the crisis, thus tying
the hands of the city's financial managers even more. New York Times, April 4, 1976, p. E6;
April 18, 1976, p. E6.
18. The seminal work here is Robert Rosenthal and Lenore Jacobson, Pygmalion in the
Clossroom (New York: Hoit, Rinehart and Winston, 1968).
19. Martin Rein, "Welfare and Housing," Joint Center Working Papers Series, no. 4 (Cam-
bridge, Mass.: Joint Center for Urban Studies, Spring, 1971, rev. Feb. 1972).
20. On the alleged importance of bureaucratic detachment in processing clients see Peter
Blau, Exchange and Power in Social Life (New York: John Wiley, 1g64), p. 66.
21. See National Advisory Commission on Civil Disorders, Report (New York: Bantam,
1g68); Peter Rossi et al., Roots of Urban Discontent (New York: John Wiley, 1974).
22. Frances Fox Piven and Richard Cloward, Poor People's Movements (New York: Pan-
theon, 1977), pp. 2o-21.
23. Michael Lipsky and Margaret Levi, "Community Organization as a Political Resource,"
in Harlan Hahn, ed., People and Places in Urban Society (Urban Affairs Annual Review, vol. 6)
(Beverly Hills, Calif.: Sage Publications, 1972), pp. 175-199.
24. See James O'Connor's discussion of "legitimation" and his general thesis concerning
the role of the state service sector, in O'Connor, The Fiscal Crisis of the State (New York: St.
Martin's, 1973). On social control functions in particular policy sectors see Samuel Bowles and
Herbert Gintis, Schooling in Capitalist America (New York: Basic Books, 1976); Frances Fox
Piven and Richard Cloward, Regulating the Poor (New York: Pantheon, 1971); Galper, The Poli-
tics of Social Services; Richard Quinney, Criminology (Boston: Little, Brown, 1975); Ira Katznel-
son, "Urban Counterrevolution," in Robert P. Wolff, ed., 1g84 Revisited (New York: Alfred
Knopf, 1973), pp. 139-164.
Notes
Chapter 2
1. See Chris Argyris, Integrating the Individual and the Organization (New York: John
Wiley, 1g64), pp. 35-41.
2. Frank L. Morris, Sr., "The Advantages and Disadvantages of Black Political Group Ac-
tivity in Two Northern Maximum Security State Prisons" (Ph. D. diss., Massachusetts Institute
ofTechnology, 19]6), p. 40.
3 For some analysts the defining characteristic of professionalism is simply the discretion
to make decisions about clients. In this view street-level bureaucrats would unquestionably be
professionals. See Albert Reiss, The Police and the Public (New Haven: Yale University Press,
1971), p. 122.
4 On rules and the police, see James Q. Wilson, Varieties of Police Behavior (Cambridge:
Harvard University Press, 1g68), p. 31; David Perry, Police in the Metropolis (Columbus, Ohio:
Charles Merrill, 1975), p. 168. See also Gresham Sykes' discussion of the dilemma of prison
guards in being formally required to intervene in all cases of observed infractions, in Sykes, The
Society of Captives (Princeton, N.J.: Princeton University Press, 1958).
5 For example, the Office of Civil Rights of the Department of Health, Education, and
Welfare has responsibility to monitor potential violations as follows: racial discrimination under
Title VI of the Civil Rights Act of 1964 in 16,000 public school districts, 2,8oo institutions of
higher education, and 30,000 institutions of health and social services; in the same areas, dis-
crimination against handicapped people under Section 504 of the Vocational Rehabilitation Act
of 1973; sex discrimination under Section 799A of the Public Health Service Act in 1, 500 health
education institutions, and under Section 745, sex discrimination in nursing schools; sex dis-
crimination under Title IX, Education Amendments of 1972, in 16,000 public school districts;
discrimination by federal contractors under Executive Order u246, innumerable contractors at
863 higher-education campuses, and more than 3,500 additional locations. Virginia Balderama,
"The Office of Civil Rights as a Street-Level Bureaucracy," unpublished seminar paper, Uni-
versity of Washington, March, 19]6.
6. David Perry and Paula Sornoff report that welfare workers' behavior with clients in Cali-
fornia is ruled by ll5 pounds of regulations; that the average police officer is obliged to enforce
approximately 30,000 federal, state, and local laws. Perry and Sornolf, "Street Level Adminis-
tration and the Law: The Problem of Police Community Relations," Criminal Law Bulletin, vol.
8, no. 1 (January-February, 1972), p. 46.
7 Consider police assertions that they would be less willing to risk intervention if civilian
review boards could penalize them for errors of judgment made under hectic and confusing cir-
cumstances which civilians might not appreciate.
8. For a discussion of attempts to introduce uniform sentencing for juvenile offenders, see
the report of the findings of the Juvenile Justice Standards Project, New York Times, November
30, 1975, p. 1; for adult offenders, New York TimBs, October 16, 1977, p. 1.
9 See James Q. Wilson, "The Bureaucracy Problem," The Public Interest (Winter, 1g67),
pp. 3-9
10. Keith Stevenson and Thomas Willemain, "Analyzing the Process of Screening Calls for
Emergency Service" (Cambridge, Mass.: Operations Research Center, Massachusetts Institute
of Technology, September, 1974), Technical Report TR-oB-74.
u. Interviews with administrative personnel, Veterans Administration hospital, BedfOrd,
Mass., August, 1974.
12. Fred Hechinger, "Where Have All the Innovations Gone?" New York Times, No-
vember 16, 1975, p. ED3o.
13. The emphasis here is on structural explanations. Lower-level participants may also per-
sonally disagree with policy objectives. See Donald Van Meter and Carl Van Horn, ''The Policy
Implementation Process: A Conceptual Framework," Administration and Society, vol. 6, no. 4
(1975). pp. 482-483.
14. For this analysis I have drawn on Daluendorf' s observation that assuming ubiquitous
conftict among social units helps in understanding some political events better than assuming
inclinatiq!is toward stability, integration, and interdependence. For Daluendorf, conftict rela-
tions are inevitable since authority relations, which are present in all social units, are necessarily
Notes
relations of subordination and superordination. However, Dahrendorf for general purposes is
unable to choose between the two models of social dynamics-the integration or the "coercion"
model-although for an analysis of class formation and development he favors the coercion
perspective. Similarly, the model stressing conflict outlined here may be applicable under the
circumstances outlined here, while the systems-integration model may be appropriate for other
aspects of policy analysis. Generally see Ralf Dahrendorf, Class and Class Conflict in Industrial
Society (Stanford, Calif.: Stanford University Press, 1959), chap. 5
15. The perspective developed in these paragraphs is elaborated in Michael Lipsky,
"Standing the Study of Public Policy Implementation on Its Head," in W. Dean Burnham and
Martha Wagner Weinberg, eds., American Politics and Public Policy (Cambridge, Mass.: Mas-
sachusetts Institute of Technology Press, 1978), pp. 391-402.
16. Argyris, Integrating the Individual and the Organization, pp. 5g-67.
17. On low motivation of public service workers, see Eric Nordlinger, Decentralizing the
City (Cambridge, Mass.: Massachusetts Institute ofTechnology Press, 1972), chap. 3; and E. S.
Savas and Sigmund Ginsburg, "The Civil Service-A Meritless System?" The Public Interest,
no. 32 (Summer, 1973), pp. 7o--85.
The problem of maintaining worker participation in organizations is a classic issue of organi-
zational theory. For a significant early analysis, see James March and Herbert Simon, Organiza-
tions (New York: John Wiley, 1958).
18. Dahrendorf, Class and Class Conflict in Industrial Society, p. 178.
19. Donald Van Meter and Carl Van Horn point out that the "disposition of implementors"
is critical to policy implementation success. The following discussion elaborates two of the con-
ditions under which, they assert, policy implementors will resist implementation: when the
policies to be implemented offend their sense of self-interest, and when the policies threaten
features of the organization and procedures they desire to maintain. Van Meter and Van Horn,
"The Policy Implementation Process: A Conceptual Framework," pp. 482-483.
20. The discussion is necessarily schematic to a degree. For example, it is an over-
simplification to treat street-level bureaucracies as comprised of lower-level workers and man-
agers. In this discussion the term "manager" refers to someone in an immediate supervisory
position vis-a-vis street-level bureaucrats (for example, a supervisor in a public welfare agency,
a police captain in charge of a precinct sector, or a principal in a nondepartmentalized public
school). "Objectives" refers to the goals that the supervisor is charged with realizing. It is neces-
sary to put it this way because the role of supervisor is itself subordinate to other roles in a
complex bureaucracy. The focus on the divergence of objectives between the organization and
the lowest-level workers could with some modifications be applied to the relations between the
lowest-level supervisor and the roles to which this position is subordinate.
21. Argyris, Integrating the Individual and the Organization, p. 36.
22. For an extended treatment of the sources of street-level bureaucrats' influence, see
Jeffrey Prottas, People-Processing: The Street-Level Bureaucrat in Public Service Bureaucracies
(Lexington, Mass.: Lexington Books, 1979).
23. This paragraph is based upon personal observations, conversations with court person-
nel, and sustained discussions with workers in the Boston Court Resources Project.
24. See Jeffrey Prottas, People-Processing, chapter 3
25. See Jon Pynoos, "Breaking the Rules: The Failure to Select and Assign Public Housing
Tenants Equitably," (Ph.D. diss., Harvard University, 1974).
26. Internal memo, "Budget Bureau Recommendations for Saving in the Welfare Budget,"
March 24, 1g6g, p. IV--6 (author's files).
27. Kuh said he acted under a general provision of the law permitting prosecutors to use
discretion to assure humane and rational dispositions. See New York Times, June 19, 1974, p. 1.
At this date 87 methadone "sellers" were affected by his decision.
28. Pietro Nivola, "Municipal Agency: A Study of the Housing Inspectional Service in Bos-
ton," (Ph.D. diss., Harvard University, 1976, chap. 7).
29. David Mechanic, "Sources of Power of Lower Participants in Complex Or~~;anizations,"
Administrative Science Quarterly, vol. 7, no. 2 (December, 1g6z), pp. 349-364.
30 Ibid., p. 352
31. See Jonathan Rubinstein, City Police (New York: Farrar, Strauss, Giroux, 1973), chap.
2.
Notes
32. Thomas Scheff, "Control over Policy by Attendants in a Mental Hospital," journal of
Health and Human Behavior, vol. 2(1glh), p. 97, cited in Mechanic, "Sources ofPower," p. 363.
Introduction
1. For vivid descriptions of street-level bureaucracies from the client perspective, see Paul
Jacobs, Prelude to Riot (New York: Vintage, 1g68); Joseph Lyford, The Airtight Cage (New York:
Harper & Row, 1g66).
2. Two recent studies of policy making focusing on the importance of the context of decision
making, particularly lack of resources and uncertainty, are Martha Wagner Weinberg, Manag-
ing the State (Cambridge, Mass.: Massachusetts Institute ofTechnology Press, 1977); Douglas
Yates, The Ungovernable City (Cambridge, Mass.: Massachusetts Institute ofTechnology Press,
1977).
3 In this discussion the word "client" is used to refer to the subjects of interactions of
street-level bureaucrats. This creates several problems in comparison with common usage. For
example, the word "client" commonly is used to refer to people for whom service is performed.
In this sense the clients of police are the people (or general public) being protected, rather than
the subjects of interaction, which include robbers as well as robbed. Also, there are generic
words for the subjects of service which make "client" appear awkward (e.g., doctors' patients,
teachers' students or pupils). However, considering all the difficulties it seems less pedantic to
refer to "clients" rather than "subjects" and truer to the synthetic objectives of this study to
refer often to "clients" rather than to utilize in general discussion the generic words for sub-
jects of study. I trust the reader will bear with me on this point.
For a discussion of the implications of various terms designating the lowest levels of organiza-
tional participation see Amitai Etzioni, A Comparative Analysis ofComplex Organizations (New
York: The Free Press, 1g61), pp. 17-21.
Chapter 3
243
Notes
6. See John H. McNamara, "Uncertainties in Police Work: The Relevance of Police Re-
cruits' Backgrounds and Training," in David Bordua, The Police: Six Sociological Essays (New
York: John Wiley, 1g67), pp. 168-177-
7 Maureen Mileski, "Courtroom Encounters: An Observation Study of a Lower Criminal
Court," Law and Society Review, vol. 5, no. 5 (May, 1971), p. 479
8. See Zimmerman, "The Practical Basis of Work Activities in a Public Assistance Organi-
zation."
g. Richard Weatherley points out that paperwork also protects workers from clients and
provides solace from job pressures. Many workers appreciate the required interruptions from
seeing clients, and thus many actually depend upon paperwork routines as job-coping devices
that help moderate the work day.
10. For a good discussion of the transition from rookie to veteran, see John Van Maanen,
"Working the Street: A Developmental View of Police Behavior," in Herbert Jacob, ed., The
Potential for Reform of Criminal justice (Beverly Hills, Calif.: Sage, 1974).
11. See James Q. Wilson, Thinking About Crime (New York: Basic Books, 1975), chaps. 3.
8.
12. See Wilson, Varieties of Police Behavior (Cambridge, Mass.: Harvard University Press,
1g68), pp. 1g-2o.
13. Carl Werth man and Irving Piliavin, "Gang Members and the Police," in Bordua, ed.,
The Police: Six Sociological Essays, p. 74
14. See, for example, William A. Westley, "Violence and the Police," American journal of
Sociology, vol. 59 (August, 1953), p. 39; Werthman and Piliavin, "Gang Members and the
Police," p. 93; Richard Blum, "The Problems of Being a Police Officer," Police (January, 1961)
p. 12.
15. Gresham Sykes, The Society of Captives (Princeton, N.J.: Princeton University Press,
1958).
16. Georgette Bennett-Sandier and Earl Ubell, "Time Bomb in Blue," New York, March
21, 1977. p. 47
17. Alfred M. Bloch, "The Battered Teacher-A New Form of Combat Neurosis," un-
published paper dated March 27, 1976.
18. Howard Becker, "Social Class and Teacher-Pupil Relationships," in Blaine Mercer and
Edwin Carr, eds., Education and the Social Order (New York: Holt, Rinehart and Winston
1957), pp. 278-z7g; Bernard Kelner, How to Teach in Elementary School (New York: McGraw-
Hill, 1958), p. 19.
19. B. L. Margolis et al., "Job Stress: An Unlisted Occupational Hazard," journal of Oc-
cupational Medicine, vol. 16, no. 10 (Oct., 1974), pp. 65g-661. Interestingly, the most consis-
tent relationships between mental health and working conditions occurred with respect to two
other indicators: underutilization of workers' abilities and nonparticipation in decisions affect-
ing one's job.
20. George Kirkham, "What a Professor Learned When He Became a Cop," U.S. News
and World Report, April 22, 1974. p. 72.
21. In the early 1970s the number of emergency room visits incr~ased at the rate of about
10 percent per year. Thomas Willemain, "The Status of Performance Measures for Emergency
Medical Services," Technical Report No. o6-74 (Cambridge, Mass: Operations Research Cen-
ter, Massachusetts Institute ofTechnology, July, 1974), p. 3 This figure is clearly too large to be
accounted for by increases in population or the absolute number of emergencies experienced.
According to a study of Chicago emergency facilities the most important factor in accounting for
an increase in utilization of 82 percent per 1,000 population from 1g6o to 1g6g was the increase
in the number of people using the emergency rooms for treatment of nonemergency conditions.
Barry Schwartz, Queuing and Waiting (Chicago: University of Chicago Press, 1975), p. 127.
22. Catherine Kohler Riessman, "The Supply-Demand Dilemma in Community Mental
Health Centers," American journal of Orthopsychiatry, vol. 40, no. 5 (October, 1970), pp.
8s8-86s
23. This is the hypothesis of Richard A. Posner, "An Economic Approach to Legal Proce-
dure and Judicial Administration," journal of Legal Studies, vol. 2 (1973), pp. 447-448. Cf.
Levin, "Delay in Five Criminal Courts," pp. 127-128.
24. Robert Perlman, Consumers and Social Services (New York: John Wiley, 1975), p. 70.
2!'). For discussions of different aspects of changing expectations of the police see Wilson,
244
Notes
Varieties of Police Behavior; Arthur Waskow, From Race Riot to Sit-in (Garden City, N.Y.:
Doubleday, 1g66); James Richardson, "To Control the City: The New York Police in Historical
Perspective," in Kenneth T. Jackson and Stanley Schultz, eds., Cities in American History (New
York, Knopf, 1972), pp. 2&r-288; Allan Silver, "The Demand for Order in Civil Society: A
Review of Some Themes in the History of Urban Crime Police and Riot," in David Bordua, ed.,
The Police, pp. 1-24; Robert Fogelson, Big City Police (Cambridge, Mass: Harvard University
Press, 1977).
26. C. H. Goodrich et. a!., "The New York Hospital-Cornell Medical Center: Progress
Report on an Experiment in Welfare Medical Care," American Journal of Public Health, vol.
55, no. 1 (1g65), pp. 88-g3; James Weiss and Merwyn Greenlick, "Determinants of Medical
Care Utilization: The Effect of Social Class and Distance on Contacts with the Medical Care
System," Medical Care, vol. 9 (1970). Cited in Deborah Stone, Institute of Policy Sciences,
Duke University, "Professionals and Social Services," unpublished paper (March, 1976).
27. Carol Ruth Silver, "The Imminent Failure of Legal Services fur the Poor: Why and
How to Limit Caseload," Journal of Urban Law, vol. 46 (1g6g), p. 217.
28. For a similar analysis applied to collective demands see Michael Lipsky and David J.
Olson, Commission Politics: The Processing of Racial Crisis in America (New Brunswick, N.J.:
Transaction Books, 1977), pp. 3-6.
29. Nordlinger makes a well-reasoned argument that out of eighty thousand complaints
concerning city services registered in Boston in 1970, fully fifty to sixty thousand would not have
been registered in the absence of the Little City Halls program. He estimates that at least half of
these calls, representing approximately a third of these service complaints, were legitimate and
not crank complaints or trivial. In other words, at least a third of all the service complaints
received were new, yet of the kind the city had been receiving under old demand-receiving
policies. See Eric Nordlinger Decentralizing the City: A Study of Boston's Little City Halls
(Cambridge, Mass.: Massachusetts Institute ofTechnology Press, 19722, p. 286.
30. Reissman, fur example, cites the case of an Oklahoma public health clinic that increased
tenfold the number of persons brought to the clinic for immuniZation by relying for outreach on
seven paraprofessionals rather than three nurses. Reissman, "The Demand-Supply Dilemma,"
p. 858.
31. David Kirp, "Schools as Sorters: The Constitutional and Policy Implications of Student
Classification," University of Pennsylvania Law Review, vol. 121, no. 4 (April, 1973), p. 712.
32. Some of the inactive caseload may consist simply of cases that would not be on the rolls
if the worker had had time to find out that they should be eliminated fur reasons of changed cir-
cumstances. This enrages agency administrators, particularly in welfare or other entitlement
programs. For a discussion of these caseload dynamics in legal services see Hosticka, "Legal
Services Lawyers Encounter Clients."
33 Here is a concrete if hypothetical illustration. If, on the average, legal service lawyers
have formal case loads of eighty and active case loads of twenty, if another lawyer were added to
a five-person office, and if no new cases were accepted, then each attorney would have approxi-
mately 66 cases (the old case load of the office now divided by six). But each attorney would still
have an active case load of twenty. This increases the number of clients actively served, but
does not improve the situation qualitatively. The pressure of time remains unchanged since the
active case load is a function not of work assigned but of the amount of case pressure that work-
ers can accommodate.
34 Generally, see chap. 11.
35 For example, Milton Heumann has fOund that a loss of adversary activity in the courts
and an increase in plea bargaining apparently are not directly related to case-load pressures. See
Heumann, "A Note on Plea Bargaining and Case Pressure," Law and Society Review, vol. 9, no.
3 (Spring, 1975), PP 515-528.
36. Ibid., p. 527
37 See the discussion in Robert Alford, Health Care Politics (Chicago: University of Chi-
cago Press, 1975), p. 222.
38. For a discussion of costs imposed on clients, see chapter 8 herein.
39 See Downs, Inside Bureaucracy, p. 188.
245
Notes
Chapter 4
1. On goal conflicts in police work see James Q. Wilson, Varieties of Police Behavior
(Cambridge, Mass.: Harvard University Press, 1g68); in public education, see Jeffrey Raffel,
"Responsiveness in Urban Schools: A Study of Adaptation to Parental Preferences in an Urban
Environment," (Ph.D. diss., Massachusetts Institute of Technology, 1972); in public welfare,
see Gilbert Steiner, The State of Welfare (Washington, D.C.: Brookings, 1971).
2: Willis Hawley, "Dealing with Organizational Rigidity in the Public Schools," (paper pre-
sented at the Annual Meeting of the American Political Science Association, September, 1971),
p. 22, n. 77 See also Yeheskel Hasenfeld and Richard English, eds., Human Service Organiza-
tions (Ann Arbor, Mich.: University of Michigan Press, 1974), pp. g-12.
3 Martin Landau, "On the Concept of a Self-Correcting Organization," Public Administra-
tion Review, vol. 33, no. 6 (November-December, 1973), p. 536.
4 Daniel P. Moynihan, Maximum Feasible Misunderstanding (New York: The Free Press,
1g6g), chap. 5
5 Martin Rein, Social Policy (New York: Random House, 1970), p. xi.
6. Hasenfeld and English, Human Service Organizations, pp. 12-14.
7 See Gary Bellow and Jeanne Kettleson, "From Ethics to Politics: Confronting Scarcity
and Fairness in Public Interest Practice," Boston University Law Review, vol. 58, no. 3 (May,
1978), pp. 337-390
8. Amitai Etzioni, A Comparative Analysis of Complex Organizations (Glencoe, Ill.: Free
Press, 1g61),
9 Because teachers, social workers, nurses, and other occupational groups typically display
professional characteristics such as length of training, degree of autonomy, etc., to a lesser
degree than the professions of medicine and law, some analysts have chosen to call them "semi-
professions" to highlight this distinction. See Amitai Etzioni, ed., The Semi-Professions and
their Organization (New York: The Free Press, 1g6g).
10. Martin Levin, "Delay in Five Criminal Courts," journal of Legal Studies, vol. 4, no. 1
(January, 1975), P go.
11. On the tendency for organizations to maintain themselves and enhance their position
see, notably, Chester Barnard, The Functions of the Executive (Cambridge, Mass.: Harvard
University Press, 1938).
12. See then President Gerald Ford's statement proposing to reduce the runaway risks of the
food-stamp program, New York Times, October 10, 1975.
!3. Michael Lipsky and Morris Lounds, "Citizen Participation in Health Care: Dilemmas
of Government Induced Participation," journal of Health Politics, Policy and Law, vol. 1, no. 1
(Spring, 1976), pp. 85-111.
14. Theodore Sarbin and Vernon Allen, "Role Theory," in Gardner Lindzey and Elliot
Aronson, eds., The Handbook of Social Psychology, 2d ed. (Reading, Mass.: Addison-Wesley,
lg68), PP 488-s67. esp. PP 4g8-499, 532.
15. See Wilson, Varieties of Police Behavior; Raffel, "Responsiveness in Urban Schools";
and Steiner, The State of Welfare.
16. On differences in cities' political cultures see Robert Alford, Bureaucracy and Partici-
pation: Political Culture in Four Wisconsin Cities (Chicago: Rang-McNally, 1g6g), and Herbert
Jacob, Debtors in Court: The Consumption of Government Services (Chicago: Rand McNally,
1g6g).
17. Wilson discusses the "zone of indifference" in which police administrators are free to
act in Varieties of Police Behavior, p. 233. The phrase is from Barnard, The Functions of the Ex-
ecutive, p. 167.
18. A case in point was provided by Boston policemen charged with preventing white Bos-
ton residents from harassing black school children when Boston schools were integrated in
1975. The policemen often grew up in the same neighborhoods as those in the crowds they were
trying to control. They were friendly with or lived among neighborhood residents and did not
personally approve of school integration as it was being carried out. See John Kifner, "Men in
the Middle," New York Times Magazine (September 12, 1976), pp. 36ff.
19. The theme of role conflict pervades the literature on police. See, for example, Carl
Notes
Werthman and Irving Piliavin, "Gang Members and the Police," Albert Reiss and David Bor-
dua, "Environment and Organization: A Perspective on the Police," and James Q. Wilson,
"Police Morale, Reform, and Citizen Respect: The Chicago Case," in Bordua, ed., The Police;
Herman Goldstein. "Police Discretion: The Ideal Versus the Real." Public Administration Re-
view, vol. 23 (September, 19&.3), p. 142; Arthur Niederhoffer, Behind the Blue Shield (New
York: Doubleday, 1g67).
20. Perhaps the extreme expression of the exclusion of clients as a reference group is found
in the courts. As Martin Levin writes, the person "with perhaps the most potential interest in
the criminal court-the victim-usually does not even watch its proceedings, and when he
does, he does not exercise effective supervision. Indeed, almost all aspects of the court .vrocess
. . . operate to discourage his effort to watch." Levin, "Delay in Five Criminal Courts, p. 95
2.1. See Norman Fainstein and Susan Fainstein, Urban Political Movements (Englewood
Cliffs, N.J.: Prentice-Hall, 1974).
22. Research findings in this area are summarized in Sarbin and Allen, "Role Theory," pp.
so:rso6. In their study of job stress Margolis et al. found role ambiguity to be associated with
six out of ten measures of job stress. "Job Stress: An Unlisted Occupational Hazard," pp.
65g-661.
2.3. Downs, Inside Bureaucracy, chap. 3
2.4 For further discussion of problems in measuring performance, see chapter 11. Here the
objective is simply to elaborate the proposition that unavailability of performance measures is
a common, critical condition of street-level bureaucracy work.
2.5. Rubinstein, City Police, pp. 32.-43, 67.
26. John I. Kits use and Aaron V. Cicourel, "A Note on the Use of Official Statistics," Social
Problems, vol. 11 (19&.3), pp. 131-139.
2.7. Peter Blau, The Dynamics of Bureaucracy, rev. ed. (Chicago: University of Chicago
Press, 1g63), chap. 3
z8. See Stanton Wheeler, "The Structure of Formally Organized Socialization Settings," in
'Orville Brim, Jr. and Stanton Wheeler, Socialization after Childhood: Two Essays (New York:
John Wiley, 1g66), pp. 102.ff.
We may hypothesize that the willingness of street-level bureaucrats to develop, and the
public to accept, these surrogate measures of performance reinfOrces conservative tendencies.
When these reified qualities are accepted as good or significant, or even if they simply deter-
mine the reward structure of the agency, the people who display these characteristics remain
entrenched. Significantly, it is highly upsetting to the status quo when these surrogate mea-
sures are challenged. This is why reform in police departments may be enhanced by insisting
that a college degree be a condition of employment. Those who prospered under the old system
are disadvantaged by this innovation. But this would only be true in departments that pre-
viously had a relatively uneducated staff. To take the point to the extreme, it would also be a
destabilizing and possibly advantageous reform in a highly educated department to forbid
employment eX college graduates. Similarly, it might disrupt the status quo to reward teachers
who have previously had outside work experience or to reward teachers who show particular
abilities in interacting with a wide range of students, since these are qualities that are not nor-
mally rewarded by public school systems.
2.9. Hawley, "Organizational Rigidity in the Public Schools," p. 13.
30. James G. Anderson, "The Authority Structure of the School: System of Social
Exchange," Educational Administration Quarterly, vol. 3 (Spring, 1!J67), p. 136, cited in Haw-
ley, "Organizational Rigidity in the Public Schools," p. 13.
31. David Seidman and Michael Couzens, "Crime, Crime Statistics, and the Great Ameri-
can Anti-Crime Crusade: Police Misreporting of Crime and Political Pressures" (Paper pre-
sented at the Annual Meeting of the American Political Science Association, Washington, D.C.
1972.).
32. Genel'l!lly see Pietro Nivola, "Municipal Agency: A Study of the Housing Inspection
Service in Boston." (Ph.D. diss., Harvard University, 1976).
33 Albert 0. Hirschman, Exit, Voice and Loyalty (Cambridge, Mass.: Harvard University
Press, 1970).
34 See Hawley, "Organizational Rigidity in the Public Schools," p. 12..
35 Gary Bridges, "Citizen Choice in Public Services: Voucher Systems," in E. S. Savas,
ed., Alternatives for Delivering Public Services (Boulder, Colo.: Westview Press, 1977), pp.
247
Notes
51-109; David K. Cohen and Eleanor Farrar, "Power to the Parents?-The Story of Education
Vouchers," Public Interest, no. 48 (Summer, 1977), pp. 72-97.
Chapter 5
1. Others have commented on the extent to which clients of some organizations are non vo-
luntary, and they have attempted to assess the importance of this distinction for treatment and
organizational behavior. See Elaine Cumming, Systems of Social Regulation (New York: Ather-
ton, 1g68). James D. Thompson discusses the extent to which two variables (the extent to which
a member of an organization is tightly or loosely controlled by its rules and assumptions and
whether the nonmembers, in our case clients, participate in the interaction voluntarily or not)
affect the relationship between members and nonmembers of an organization. Thompson's
dichotomization of the interaction variable presents the choices in extreme form and does not ac-
commodate gradations. Thompson's two variables when combined yield four organizational types;
our analysis draws attention to the probability that, analytically, the participation of most poor
people in transactions with public agencies tends to be mandatory. See Thompson, "Organiza-
tions and Output Transactions, in Elihu Katz and Brenda Danet, eds., Bureaucracy and the
Public (New York: Basic Books, 1972), pp. 191-211.
2. The extent to which clients can affect doctors' behavior is the subject of Eliot Freidson's
article, "Client and Medical Practice," American journal of SociokJgy, vol. 65 (January, 1g6o),
374-382. See also, Amitai Etzioni, "Administration and the Consumer," American SociokJgical
Quarterly, vol. 3, no. 2 (September, 1955), pp. 257-264.
3 See Willis Hawley, "Organizational Rigidity in the Public Schools," (paper presented at
the Annual Meeting of the American Political Science Association, September, 1971), p. 15.
4 I have found the most accessible discussion of bargaining to be Thomas Schelling, The
Strategy of Conflict (Cambridge, Mass.: Harvard University Press, 1g6o).
5 Julius Roth, "Some Contingencies ofthe Moral Evaluation and Control of Clientele: The
Case of the Hospital Emergency Service," in Yeheskel Hasenfeld and Richard English, eds.,
Human Seroice Organizations, (Ann Arbor, Mich.: University of Michigan Press, 1974), p. 502.
For a general orientation to the perspective implicit here see Erving Gotfman, Strategic In-
teractions (Philadelphia, Pa.: University of Pennsylvania Press, 1g6g).
6. Eliot Freidson, Professional Dominance (New York: Atherton, 1970).
7 See Gresham Sykes, Society of Captives (Princeton, N.J.: Princeton University Press,
1958), pp. 4S-s8.
8. For a discussion of ways inmates are induced to contribute to their own control in mental
hospitals and other institutions that totally circumscribe people's lives see Erving Gotfman,
Asylums (Garden City, N.Y.: Doubleday, 1g61), pp. 177-207.
9 Gotfman, Strategic Interactions.
10. Ira Katznelson, Black Men, White Cities (New York: Oxford University Press, 1973), p.
25. On the conceptualization of research on relationships of dominance and subordination, see
chap. 2.
11. The title of this section is obviously a paraphrase (apt, I trust) of Peter Berger and
Thomas Luckmann, The Social Construction of Reality (Garden City, N.Y.: Doubleday, 1976).
For a detailed elaboration of the process of client categorization, see Jeffrey Prottas, People-
Processing: The Street-Level Bureaucrat in Public Service Bureaucracies (Lexington, Mass.:
Lexington Books, 1979).
12. For the importance of these distinctions in court processing of juveniles see Robert
Emerson, judging Delinquents (New York: Aldine, 1g6g).
13. On the distinction between structure and behavior see Katznelson, Black Men, White
Cities, chap. 2.
14. Elihu Katz and S. N. Eisenstadt, "Some Sociological Observations on the Response of
Notes
Israeli Organizations to New Immigrants," in Elihu Katz and Brenda Danet, eds., Bureaucracy
and the Public, p. 79
15. Reported by Carl Hosticka, "Legal Services Lawyers Encounter Clients. A Study in
Street-level Bureaucracy" (Ph.D. diss. Massachusetts Institute of Technology, 1976).
16. Carl Werthman and Irving Piliavin, "Gang Members and the Police," in David Bordua,
ed., The Police: Six Sociological Essays (New York: John Wiley, 1g67), p. 87.
17. Michael Iipsky and Margaret Levi, "Community Organization as a Political Resource,"
in Harlan Hahn, ed., People and Politics in Urban Society (Beverly Hills, Calif.: Sage, 1972),
pp. 195-1g6; Michael Iipsky, Protest in City Politics (Chicago, Ill.: Rand McNally, 1970).
18. In coaching the client street-level bureaucrats are only contributing to a process clients
otherwise engage in: maneuvering to secure what they think will provide the best chance or the
most favorable outcome (getting the best judge, teacher, social worker; phrasing words cor-
rectly; having papers ready, etc.).
19. Jon Pynoos, "Breaking the Rules: The Failure to Select and Assign Public Housing
Tenants Equitably," (Ph.D. diss., Harvard University, 1974). (See chap. 2, pp. 21--22).
20. Pietro Nivola, "Municipal Agency: A Study of the Housing lnspectional Service in Bos-
ton," (Ph.D. diss. Harvard University, 1976), chap. 3
21. Alan Keith-Lucas, Decisions about People in Need (Chapel Hill, N.C.: University of
North Carolina Press, 1957), p. 224.
22. Jerome Skolnick describes how defense attorneys instruct clients fur best results to
express regret and perplexity at their own behavior rather than attempt to explain the behavior
away, although thi is their inclination. Skolnick, "Social Control in the Adversary System,"
]oumalofConflictResolution, vol. 11, no. 1 (1967), pp. 59-&7, in Jerome Skolnick and Richard
Schwartz, eds., Society and the Legal Order (New York: Basic Books, 1970), pp. 414-423, cita-
tion at p. 418.
23. The tendency of bureaucrats to treat some part of the population specially, in opposi-
tion to formal mandates, is explored in Brenda Danet," 'Giving the Underdog a Break': Latent
Particularism Among Customs Officials," in Katz and Danet, Bureaucracy and the Public, pp.
329-337. Danet discusses such tendencies in terms of bureaucrats' "latent particularism." From
an organizational standpoint "latent particularism'' is discussed in terms of "debureaucratiza-
tion" in Elihu Katz and S. N. Eisenstadt, "Some Sociological Observations on the Response of
Israeli Organizations to New Immigrants," in Katz and Danet, pp. 73-88. Coaching the client as
discussed here is one form of "debureaucratization." The term is meant to reflect departure
from an ideal type of bureaucratic universalism. However, the term is somewhat awkwarrl since
it implies a departure from a state of bureaucratic fOrmalism. A bureaucracy which hR never
achieved the universalism of the ideal type in common terms cannot usefully be described as
debureacratized.
24. For a general treatment of this perspective, see Goffinan, Strategic Interactions.
25. Joel Handler and Mary Jane Hollingsworth, The Deserving Poor (Chicago: Markham,
1971). See also Kenneth Clark, Dark Ghetto (New York: Harper & Row, 1g65).
26. Goffinan, Asylums; David Mechanic, Medical Sociology: A Selective View (New York:
The Free Press, 1g68), pp. 11sff.
27. Cf. Murray Edelman's view of the "helping professions" in Political Language: Words
That Succeed and Policies That Fail (New York: Academic Press, 1977), chap. 4
28. Robert Rosenthal and Lenore Jacobson, PygmaUon in the Classroom (New York: Holt,
Rinehart and Winston, 1g68); Ray C. Rist, "Student Social Class and Teacher Expectations: The
Self-Fulfilling Prophecy in Ghetto Education," Harvard Educational Review, vol. 40 (August,
1970), pp. 411-451; see Robert Merton, Social Theory and Social Structure (Glencoe, Ill.: The
Free Press, 1957), chap. 11. Self-fulfilling prophecies that result from assigning clients to ca-
tegories without necessarily being affected by the interaction with street-level bureaucrats are
discussed in chapter 10.
29. Rist, "Student Social Class and Teacher Expectations."
30. Emerson, Jucling Delinquents.
31. Richard Weatherley, Reforming Special Education: Policy Implementation from State
Level to Street Level (Cambridge, Mass.: Massachusetts Institute of Technology Press, 1979).
32 Berger and Luckmann, The Social Construction of Reality, p. 30.
33 Eliot Freidson, Profession of Medicine (New York: Dodd Mead, 1974), pp. u6ff.
249
Notes
34 Jerome Skolnick and Richard Schwartz, ''Two Studies of Legal Stigma," Social Prob-
lema, vol. 10 (Fall, 1g&.), 133-142., cited in Maureen Mileski, "Courtroom Encounters," Law
and Society Review, vol. 5. no. 5 (May, 1971), p. 4g6.
35 Handler and Hollingsworth, The Deserving Poor; Hosticka, "Legal Servies Lawyers
Encounter Clients."
Chapter 6
1. On the role of myth in public policy see Murray Edelman, Political Language (New
York: Academic Press, 1977), esp. chap. 1.
2.. A. Cicourel and J. Kitsuse, The Educational Decision Makers (Indianapolis, Ind.: Bobbs-
Merrill, 19&.3), cited in David Kirp, "Schools as Sorters: The Constitutional and Policy Implica-
tions of School Classification," University of Pennsylvania Law Review, vol. 12.1, no. 4 (April,
1973). p. 711.
3 Deborah Stone has elaborated the dilemma of physicians who are asked to act as both
advocates and overseers of the public purse, in Controlling the Medical Profession (Chicago:
University of Chicago Press, forthcoming).
4 Richard Weatherley, Reforming Special Education: Policy Implementation from State
Level to Street Level (Cambridge, Mass.: Massachusetts Institute of Technology Press, 1979).
5 On the tension between supporting and controlling clients see Elaine Cumming, Sys-
tems of Social Regulation (New York: Atherton, 1g68), pp. 6-g.
6. Jerome Skolnick, "Social Control in the Adversary System," in Jerome Skolnick and
Richard Schwartz, Society and the Legal Order (New York: Basic Books, 1970), pp. 42.1-42.2..
7 Peter Blau, Dynamics of Bureaucracy (Chicago: University of Chicago Press, 1g64).
8. Suggested by Judy Riley, "A Case Study of Street-Level Bureaucracy: Child Protective
Services," unpublished seminar paper, University of Washington, 1976.
9 For a review and discussion of political alienation as a psychological construct see Stanley
Greenberg, "Political Alienation and Political Action," in Willis Hawley and Michael Upsky,
eds., Theoretical Perspectives on Urban Politics (Englewood Cliffs, N.J.: Prentice-Hall, 1976),
pp. 176--183.
10. Here I recognize an assumption that, while the subject of considerable debate, remains
ultimately unresolvable: self-actualization, particularly tendencies toward creativity, coopera-
tion, and personal growth, is inherently a human quality that people strive to express if given
the chance and freed from pursuit of necessities. For a discussion of this orientation in organiza-
tional studies see Chris Argyris, "Some limits of Rational Man Organizational Theory," Public
Administration Review, vol. 33 (June, 1973), pp. 2.53-2.67.
For a summary of how alienation is generally utilized to describe relations of work see Fred-
erick Thayer, An End to Hierarchy! An End to Competition! (New York: Franklin Watts, 1973),
pp. 47-48; also Amitai Etzioni, The Active Society (New York: The Free Press, 1g68), chap. 2.1.
11. Etzioni, The Active Society, pp. 618-62.0.
12.. In these paragraphs I have deliberately chosen language to suggest parallels with analy-
ses of the work of industrial workers.
13. On trends in workers' control see Administr.ation and Society, vol. 7, no. 1 (May, 1975),
an issue devoted to this topic.
14. In response to these work-related problems street-level bureaucracies often do attempt
to control the nature of the clientele. This is treated in chapters 7 through 9
15. Very few studies have concentrated on the dynamics over time of routine treatment of
clients by public agencies. See Alan Keith-Lucas, Decisions about People in Need (Chapel Hill,
N.C.: University of North Carolina Press, 1957) for a study of the treatment of welfare clients in
the south in the 1950s. See Waskow, From Race Riot to Sit-In, for a study of changes in non-
routine police practice over time. For speculative study of the implications of modernization in
police departments see James Q. Wilson, "The Police and the Delinquent in Two Cities," in
Notes
Stanton Wheeler, ed., Controlling Delinquents (New York: John Wiley, 1g68). For a discussion
of developments in the bureaucratic treatment of Israeli immigrants over time see Katz and
Eisenstadt, 'The Response of Israeli Organizations to New Immigrants," in Elihu Katz and
Brenda Danet, eds., Bureaucracy and the Public (New York: Basic Books, 1972), pp. 73--88.
Introduction
1. Cf. James Q. Wilson, "The Bureacucracy Problem," Public Interest, no. 6 (Winter,
1g67). pp. 3~
2. For another analysis that assumes people "desire to do a good job," see Downs, Inside
Bureaucracy (Boston: Little Brown, 1g67), p. 1g8.
3 Lee Rainwater, ''The Revolt of the Dirty Workers,"' Transaction, vol. 5. no. 1 (Nov.,
1g(ry-), pp. 2ff.
4 James March and Herbert Simon, Organizations (New York: John Wiley, 1958); Charles
Lindblom, "The Science of'Muddling Through'," Public Administration Review, vol. 19 (Spring
1959). pp. 7g--88.
5 Useful conceptual distinctions for various phenomena related to coping are provided by
Richard Lazarus in Psychological Stress and the Coping Process (New York: McGraw-Hill,
1g66), chap. 1.
6. Routines are the regularized or habitual patterns by which tasks are performed. Simplifi-
cations are symbolic constructs in terms of which decisions about potentially complex phenom-
ena are made, utilizing a smaller set of cues than those presented by the phenomena. Routines
are behavioral patterns of response; simplifications are mental patterns of ordering data with
which routines may or may not be associated. Peter Berger and Thomas Luckmann discuss the
ubiquitous nature of routinization and simplification in The Politics of Everyday Life (Garden
City, N.Y.: Doubleday, 1976), pp. 28ff and 53if. They use the terms "habituations" for the for-
mer, "typifications" for the latter.
7 See Reinhard Bendix, Work and Authority in Industry (New York: John Wiley, 1956),
chap. 4 Willis Hawley presents an interesting discussion of the significance of routinization
in 'The Possibilities of Nonbureaucratic Organization," Willis D. Hawley and David Rogers,
eds., Improving the Quality of Urban Management (Beverly Hills, Cali: Sage, 1974), pp.
371-.p6.
8. See Victor Thompson, Modem Organization (New York: Knopf, 1g61), p. 14; James
March and Herbert Simon, Organizations (New York: John Wiley, 1958), p. 142.
9 See the discussion of"Categorization of Data" in Thompson, Modem Organizations, p.
17
10. The importance of routines in developing policy in other areas is well established. For a
concise general treatment, see Ira Sharkansky, The Routines of Politics (New York: Van Nos-
trand, 1970).
11. On the meaning of the term "political" see David Easton, A Framework for Political
Analysis (Englewood Cliffs, N.J.: Prentice-Hall, 1g65), p. so; and Harold Lasswell, Who Gets
What, When, How? (New York: McGraw Hill, 1936).
12. Karen Orren, Corporate Power and Social Change (Baltimore: Johns Hopkins, 1973).
13. James Davis and Kenneth Dolbeare, Little Groups of Neighbors (Chicago: Markham,
1g68).
14. See Thompson, Modem Organizations, pp. 168--16g.
15. Julius Roth, "Some Contingencies of the Moral Evaluation and Control of Clientele:
The Case of Hospital Emergency Services," in Yeheskel Hasenfeld and Richard English, eds.,
Human Service Organizations (Ann Arbor, Mich.: University of Michigan Press, 1974), p. 500,
italics omitted.
Notes
Chapter 7
1. The reactive nature of police work, and police dependence upon citizens in this respect,
is stressed in Albert Reiss, The Police and the Public (New Haven: Yale University Press,
1971).
2. The latter case is cited by Barry Schwartz, Queuing and Waiting (Chicago: University of
Chicago Press, 1975), p. 24 This excellent volume provides many insights into issues of priori-
ties in client treatment and the costs of seeking service.
3 See Robert Dalll, "The Analysis oflnftuence in Local Communities," in Charles Adrian,
ed., Social Science and Community Action (East Lansing, Mich.: Michigan State University
Press, 1g6o), p. 32
4 See generally Jonathan Rubinstein, "Suspicions," in City Police (New York: Farrar,
Straus, 1973).
5 For example, one prosecutor's office that switched from telephone to mail complaint
handling in processing white collar crimes experienced a 25 percent reduction in complaints
received. Michael Brintnall, "The Allocation of Services in the Local Prosecution of Economic
Crime" (Ph.D. diss., Massachusetts Institute of Technology, 1977), chap. 6.
6. See Schwartz, Queuing and Waiting, chap. 6.
7 When court clerks use confusing legal language we may call it "bureaucratic language as
incantation." Edelman, Political Language: Words that Succeed and Policies that Fail (New
York: Academic Press, 1977), p. g8. But what shall we call the court clerk's chant that strings
words together indistinguishably? Perhaps it should be called "incantation as symbolic lan-
guage." For attempts to deal positively with the rationing effects oflegallanguage, consider the
New York state law requiring consumer contracts to be written in clear, understandable lan-
guage. See New York Times, Aug. 11, 1977, p. B1.
8. This is a paraphrase of the definition of demands in David Easton, A Framework for Po-
litical Analysis (Englewood Cliffs, N.J.: Prentice-Hall, 1g65), p. 120.
9 New York Times, September 25, 1977.
10. Leon Mayhew, "Institutions of Representation: Civil Justice and the Public," Law and
Society Review, vol. 9, no. 3 (Spring, 1975), p. 403. The discrepancy is so great that it would be
difficult to attribute it to differences in the nature of the sample.
11. Richard Cloward and Frances Fox Piven, "A Strategy to End Poverty," The Politics of
Turmoil (New York: Vintage, 1975), pp. 8g--1o6.
12. Gilbert Steiner, The State of Welfare (Washington, D.C.: Brookings, 1971).
13. Kitsuse and Cicourel have written that statistics reflect a great deal about the organiza-
tions collecting the statistics. John Kitsuse and Aaron Cicourel, "A Note on the Uses of Official
Statistics," Social Problems, vol. 11 (1963), pp. 131-139. Sometimes the statistics collectors are
not the same as those formally charged with providing information about services.
14. Eric Nordlinger, Decentralizing the City (Cambridge, Mass.,: Massachusetts Institute
ofTechnology Press, 1972), p. 286.
15. The dynamics of this process are discussed in Michael Lipsky and Morris Lounds,
"Citizen Participation and Health Care: Problems of Government Induced Participation," Jour-
nal of Health Politics, Policy and Law, vol. 1, no. 1 (Spring, 1976), pp. 85-111.
16. See the discussion of the psychological implications of waiting in Schwartz, Queuing
and Waiting, chaps. 1, 8.
17. Virtually every commentary on welfare practices draws attention to the degradation of
clients. See Alan Keith-Lucas, Decisions about People in Need (Chapel Hill, N.C.: University of
North Carolina, 1957); Steiner, The State of Welfare (Washington, D.C.: Brookings, 1971);
Piven and Cloward, Regulating the Poor (New York: Pantheon, 1971), chaps. 4-5.
18. Jeffrey Prottas, People-Processing: The Street-Level Bureaucrat in Public Service Bu-
reaucracies (Lexington, Mass.: Lexington Books, 1979). On the continuing relationships be-
tween ghetto fathers who have deserted and their families, see Elliot Liebow, TaUy's Corner
(Boston: Little, Brown, 1g67).
19. June Grant Wolf, 'The Initial Evaluation at a Walk-In Clinic: Applicant's and Evalua-
tor's Perspectives" (Ph.D. diss., Boston University, 1974), p. 76.
Notes
20. First-come, first-served, "constitutes the normative basis for most forms of queueing."
Schwartz, Queuing and Waiting, p. 93
21. Ibid., chap. 6
22. Carl Hosticka, "~Services Lawyers Encounter Clients: A Study in Street-Level Bu-
reaucracy" (Ph. D. diss., Massachusetts Institute of Technology, 1976).
23. Catherine Kohler Reissman, "The Supply-Demand Dilemma in Community Mental
Health Centers," American Journal of Orthopsychiatry, vol. 40. no. 5 (October, 1970), p. 86o.
24. See chap. 2.
25. See Jeffry Galper, The Politics of Social Services (Englewood Cliffs, N.J.: Prentice-
Hall, 1975), pp. 70-71.
26. Reissman, "The Supply-Demand Dilemma," p. 86o.
27. See Schwartz, Queuing and Waiting, pp. 2&-29.
28. Ibid., chap. 5 and fu. 5. p. 201.
29. Robert Perlman, Consumers and Social Services (New York: John Wiley, 1975), p. 77
30. Speaker, Annual Convention of the National Legal Aid and Defenders Association,
Seattle, Washington, November, 1975.
31. Nivola, "Municipal Agency: A Study of Housing lnspectional Service in Boston," chap.
3
32 "Budget Bureau Recommendations for Savings in the Welfare Budget," March 24,
1g6g. Unpublished document in author's files.
33 Ibid.
34 Ibid.
35 Ibid.
36. New York Times, December 21, 1977
Chapter 8
253
Notes
11. Robert Scott, "The Selection of Clients by Social Welfare Agencies: The Case of the
Blind," in Hasenfeld and English, Human Seroice Organizations, pp. 485-498; Donald Schon,
'The Blindness System," Public Interest, vol. 18 (Winter, 1970), pp. 25-:38.
12. LDis Forer, Death of the Law (New York: McKay, 1975).
13. Sociologist Jules Henry called the tendency of teachers to concentrate on only a few
students "partial withdrawal." See Henry, "White Peoples Time, Colored Peoples Time,"
Transaction, vol. 2, no. 3 (March-April, 1g65), p. 32
14. See New York Times, March 3, 1977, p. 33
15. Sudnow, "Normal Crimes." See also Erving Goffinan, Relations in Public (New York:
Basic Books, 1971), chap. 6.
16. Maureen Mileski, "Courtroom Encounters," Law and Society Review, vol. 5, no. 5
(May, 1971), p. 513.
17. Carl Hosticka, "Legal Services Lawyers Encounter Clients: A Study in Street-Level
Bureaucracy," (Ph.D. diss., Massachusetts Institute ofTec~nology, 1976).
18. Mileski, "Courtroom Encounters," p. 503.
19. Eliot Freidson, Profession of Medicine (New York: Dodd, Mead, 1970), p. 257
20. Children's Defense Fund, Children Out of School in America (Washington, D.C.:
Children's Defense Fund, 1974).
21. Carl Werthman and Irving Piliavin, "Gang Members and the Police," in David Bordua,
ed., The Police (New York: John Wiley, 1g67), p. 76.
22. Richard Quinney, Criminology (Boston: Little, Brown, 1975), chap. 6; Forer, Death of
the Law, chap. 5.
23. David Mechanic, Medical Sociology (New York: Free Press, 1g68), pp. usff.
24 See chap. 5 herein, note 28.
25. For a convincing discussion of the subjective bases of doctors' views of their work, see
Freidson, Profession of Medicine, pp. 168-172. The subjectivity of teachers' views of ghetto
students is suggested in Peter Rossi et al. The Roots of Urban Discontent (New York: John
Wiley, 1974), P 355
26. For a balanced discussion of the relationship between racial prejudice and generally
biased behavior in judicial sentencing see Willard Gaylin, Partial Justice: A Study of Bias in
Sentencing (New York: Vintage Books, 1975), chap. 3
Chapter g
1. Settings may also be designed to encourage client penetration of these barriers. Deut-
scher describes the case of the public-housing applications officer who was pleased to have the
receptionist's desk to one side so that prospective applicants would come directly to her desk,
bypassing the receptionist. Irwin Deutscher, 'The Gatekeeper in Public Housing," in Irwin
Deutscher and Elizabeth J. Thompson, eds. Among the People: Encounters with the Poor (New
York: Basic Books, 1g68), p. 49
2. See Stanton Wheeler, 'The Structure of Formally Organized Socialization Settings," in
OrviUe Brim, Jr. and Stanton Wheeler, Socialization after Childhood: Two Essays (New York:
John Wiley, 1g66), p. g8.
3 Ibid., p. g8. Wheeler observes that adult socialization agencies implicitly conspire to
present the client process as benign. Sometimes the conspiracy is not so implicit.
4 Street-level bureaucracies must continually justify themselves not only to a client public,
but also to a public constituency concerned with bureaucratic efficiency and effectiveness. To
carry out these tasks street-level bureaucracies expend considerable effort on public relations.
On the public relations budgets of police departments, see LDis Forer, Death of the Law, (New
York: McKay, 1975), p. 176. For an account of police affairs by a New York City deputy commis-
sioner in charge of public relations see Robert Daley, Target Blue: An Insider's View of the
N.Y.P.D. (New York: Delacorte Press, 1973).
254
Notes
5 Judy Riley, "A Case Study of Street-Level Bureaucracy: Child Protective Services," .un-
published seminar paper, University of Washington, 1976; Pietro Nivola, "Municipal Agency: A
Study of Housing Inspectional Services in Boston" (Ph.D. diss., Harvard University, 1976).
6. Carl Hosticka, "Legal Services Lawyers Encounter Clients: A Study in Street-Level Bu-
reaucracy," (Ph.D. diss., Massachusetts Institute ofTechnology, 1976).
7 Joel Handler and Mary Jane Hollingsworth, The Deseroing Poor (Chicago: Markham,
1971).
8. Howard Becker, "Social Class and Teacher-Pupil Relationships," in Blaine Mercer and
Edwin Carr, eds., Education and the Social Order (New York: Rinehart, 1957), pp. 278-279.
9 Bernard G. Kelner, How to Teach in Elementary School (New York: McGraw-Hill,
1958), p. 19. Also see Willis Hawley, "Dealing with Organizational Rigidity in Public Schools:
A Theoretical Perspective" (paper prepared for delivery at the Annual Convention of the Ameri-
can Political Science Association, September, 1971), p. 6.
10. Arthur Niederholfer, Behind the Blue Shield (New York: Doubleday, 1967), p. 53
11. Jonathan Rubinstein, City Police (New York: Farrar, Straus, 1973), pp. 301-316.
12. John Van Maanen, "Working the Street: A Developmental View of Police Behavior," in
Herbert Jacob, ed., The Potential for Reform of the Criminal Justice System (Beverly Hills,
Calif.: Sage, 1974).
13. See Jerome Skolnick, Justice without Trial (New York: John Wiley, 1g67), pp. 45-46.
See also Rubinstein, City Police, chap. 6.
14. David Kirp makes this point in an unpublished paper, "The Bureaucratization of Child-
hood."
15. Hosticka, "Legal Services Lawyers Encounter Clients."
16. Maureen Mileski, "Courtroom Encounters," Law and Society Review, vol. 5, no. 5
(May, 1971), p. 503.
17. Dennis Trees, unpublished seminar paper, University of Washington, 1976.
18. Mileski, "Courtroom Encounters," pp. 529-530.
19. James D. Thompson, Organizations in Action (New York: McGraw-Hill, 1g67), p. 123.
20. Richard Weatherley, Reforming Special Education: Policy Implementation from State
Level to Street Level (Cambridge, Mass.: Massachusetts Institute of Technology Press, 1979).
21. Nivola, "Municipal Agency."
22. Jeffrey Prottas, People-Processing: The Street-Level Bureaucrat in Public Service
Bureaucracies (Lexington. Mass.: Lexington Books, 1979), chap. 4
23. Hosticka, "Legal Services Lawyers Encounter Clients," provides a persuasive discus-
sion of the influence of receptionists. See generally, David Mechanic, "Sources of Power of
Lower Participants in Complex Organizations," Administrative Science Quarterly, vol. 7, no. 2
(December, 1g62), pp. 349-364.
24. Deutscher, "The Gatekeeper in Public Housing."
25. Joel Handler, "The Juvenile Court and the Adversary System: Problems of Function
and Form," Wisconsin Law Review, vol. 17 (Winter, 1g65).
26. Adoption-agency social workers' "nominal function of advocate for the child and advisor
to the court on the range of alternatives and possible outcomes relative to a child's development
has been transformed into a de facto assumption of judicial powers." Thomas E. Nutt and John
A. Snyder, Trans-Racial Adoption (Study supported under NIMH grant #R03 MH 1g8os-o1),
p. 19
27. Jerome Carlin, "Courts and the Poor" (paper prepared fur delivery at the 1g66 Annual
Meeting of the American Politicai Science Association, New York, September, 1g66), p. 3
28. See, e.g., Mental IUness and Due Process (Ithaca, N.Y.: Cornell University Press,
1g62), cited in Carlin, ibid.
zg. Ray C. Rist, "Student Social Class and Teacher Expectations: The Self-Fulfilling
Prophecy in Ghetto Education," in Yeheskel Hasenfeld and Richard English, eds., Human Ser-
vice Organizations (Ann Arbor, Mich.: University of Michigan Press, 1974), pp. 517-539.
30. David Sudnow, "Normal Crimes: Sociological Features of the Penal Code in a Public
Defender's Office," Social Problems, vol. 12 (Winter), pp. 255~76.
31. Julius Roth, "Some Contingencies of the Moral Evaluation and Control of Clientele:
The Case of the Hospital Emergency Room," in Hasenfeld and English, eds., Human Service
Organizations, pp. 503-504
32. Prottas, People-Processing.
255
Notes
33 An opposite practice, which is also consistent with conserving resources, is the ten-
dency to pass the buck for making determinations to other agencies. This is the current com-
plaint of prison reformers who regard indeterminate sentences, originally intended to permit
prisoners to demonstrate redeeming characteristics, as functioning to keep them under the con-
trol of prison officials who can manipulate the extension or reduction of sentence.
34 Robert Perlman, Consumers and Social Services (New York: John Wiley, 1975), p. 67.
35 Jeffry Galper, The Politics of Social Services (Englewood Cliffs, N.J.: Prentice-Hall,
1975). p. 70.
36. Rikva Bar-Yosefand E. O, Schild, "Pressures and Defenses in Bureaucratic Roles," in
Elihu Katz and Brenda Danet, eds., Bureaucracy and the Public (New York: Basic Books, 1973),
p. ~5
37 For a useful discussion of the function of appeals in the selective service system see
James W. Davis, Jr. and Kenneth Dolbeare, Little Groups of Neighbors (Chicago: Markham,
1g68), chap. 5
38. On the difficulty of filing complaints against the police see Walter Gellhorn, When
Americans Complain (Cambridge, Mass.: Harvard University Press, 1966), pp. 186fT.
39 Frances F. Piven and Richard A. Cloward, Regulating the Poor: The Functions of
Public Welfare (New York: Pantheon, 1971), p. 173. Piven and Cloward attribute the low
number of appeals to the control of the welfare system over clients, resulting in their acquies-
cence to the system of welfare on its terms, a thesis consistent with earlier arguments in this
book.
40. David C. Perry and Paula Sornoff, "Politics at the Street Level: The Select Case of
Police Administration and the Community" (rev. version of a paper presented to the Annual
Meeting of the American Political Science Association, Washington, D.C., 1972), pp. 62~3.
41. Michael Lipsky, Protest in City Politics (Chicago: Rand McNally, 1970), chap. 5
42. See, for example, David C. Perry and Paula Sornoff, "Street Level Administration and
the Law: The Problem of Police-Community Relations," Criminal Law Bulletin, vol. 8, no. 1
Ganuary-February, 1972), p. 54
43 Albert Reiss, Jr., The Police and the Public (New Haven: Yale University Press, 1971),
p. 125
44 The word "emergency" is rarely defined in studies of public services except in the spe-
cific context in which it is applied. See for example, Morris Schwartz and Charlotte Green
Schwartz, Social Approaches to Mental Patient Care (New York: Columbia University Press,
1964), p. so; Egon Bittner, "Police Discretion in Emergency Apprehension of Mentally Ill Per-
sons," Social Problems, vol. 14 (1967), pp. 278-292; Freidson, Profession of Medicine, p. u8.
45 Lipsky, Protest in City Politics, p. Bg.
46. Freidson, Profession of Medicine, p. u8.
47 See Michael Zubkoff, "Emergency Room Service," in Eli Ginsberg, ed., Urban Health
Seroices (New York: Columbia University Press, 1971), pp. ug-124
Chapter 10
257
Notes
21. Jonathan Kozol, Death at an Early Age (New York: Bantam, 1g67), pp. 10-19.
22. Kenneth Clark has analyzed theories of racial inferiority and cultural deprivation as
functional equivalents in Dark Ghetto (New York: Harper & Row, 1g65), pp. 12sff.
23. See Murray Edelman, Political Language: Words that Succeed and Policies That Fail
(New York: Academic Press, 1977).
24. Erving Coffman, Asylums (Chicago: Aldine, 1g61), pp. 86-87.
25. On differences between whites and blacks regarding ghetto residents' capabilities and
reasons for failures, see Peter Rossi et al., Roots of Urban Discontent (New York: John Wiley,
1974). On some differences in attributions of client responsibility between street-level bureau-
crats, see Clarence Stone, "Paternalism Among Social Agency Employees," Journal of Politics,
vol. 39 (August, 1977), pp. 794-8o4.
z6. Becker, "Social Class and Teacher-Pupil Relationships," pp. 278--299.
27. Judy Riley, "A Case Study of Street-level Bureaucracy: Child Protective Services" (un-
published seminar paper, University of Washington, 1976).
Chapter 11
259
Notes
in governmental effort is an empirical and normative question. Sometimes crisis can force man-
agement to attend to costs so that real savings are discovered, e.g., energy conservation by
reducing unnecessary wattage in bulbs. But at other times a change is simply justified by calling
it duplication or waste reduction although it may not be.
zz. In part, promotion and retention in street-level bureaucracies are not based on the
quality of service provision because service provision is so difficult to measure. Hence surro-
gates for effective service provision, such as tenure and advanced training, often bearing little
relationship to worker effectiveness, are used extensively to reward and promote workers.
These are not generally contradicted by more appropriate service delivery measures. On pro-
motion in street-level bureaucracies see John Van Maanen, "Working the Street: A Develop-
mental View of Police Behavior," in Herbert Jacob, ed., The Potential for Reform of the Crimi-
nal Justice System (Beverly Hills, Calif.: Sage, 1974); David Goodwin, Delivering Education
Services: Urban Schools and Schooling Policy (New York: Teachers College Press, 1977), pp.
66--67'
z3. The phrase is from Donald H. Sweet, Decruitment: A Guide for Managers (Reading,
Mass.: Addison Wesley, 1975).
z4. Letter from Hanna B. Leibowitz to New York Times, September z8, 1976, p. 38.
zs. "Frozen Means You Don't Move: The Impact of Budget Cuts on People in Mas-
sachusetts Institutions" (Massachusetts Advocacy Center, 1978), pp. 4&--47.
z6. Perhaps the most neglected aspect of the fiscal crisis is the extent to which the firing of
public employees represents a reduction in one of the critical functions of big city govern-
ments-the provision of relatively secure and decent jobs. After expressing great alarm for
many months over the fiscal crisis the New York Times eventually recognized this consideration.
"The trouble is, the bureaucracy also consists of people. Thus the fiscally sound demand for
greater economy and efficiency in the municipal health care bureaucracy could lead to the
discharge of thousands of hospital workers. In the absence of alternative job opportunities, the
result would'be suffering and despair in minority communities-and a sharp increase in welfare
rolls." November g, 1976, p. 36. Frances F. Piven has written presuasively on the redistrib-
utive aspects of urban fiscal liberalism and stringency. See an account of her views in the Boston
Globe, December g, 1976, p. 8.
z7. "Teaching is becoming an old people's profession." See ''Levittown Loses Its Younger
Teachers in Trims," New York Times, May 13, 1978, p. 27.
Chapter 12
1. V. 0. Key Jr., "Politics and Administration," in Leonard D. White, ed., The Future of
Government in the United States (Chicago: University of Chicago Press, 1942), p. 100. Cited in
Jesse McCorry, Marcus Foster and the Oakland Public Schools: Leadership in an Urban Bu-
reaucracy (Berkeley, Calif.: University of California Press, 1978), chap. 1.
z. The interplay between national culture and the organization of work is explored in
Michel Crozier, The Bureaucratic Phenomenon (Chicago: University of Chicago Press, 1g64), a
comparative study of French bureaucracy; Ronald Dore, British Factory~apanese Factory:
Origins of National Diversity in Industrial Relations (Berkeley, Calif.: University of California
Press, 1973). Arthur Stinchcombe explores the relationship of organizational structures to their
environmental origins in "Social Structure and Organization," in James March, ed., Handbook
of Organizations (Chicago: Rand McNally, 1g65), pp. 142-194 (esp. pp. 153-16g). I am in-
debted to Charles Sabel of the Massachusetts Institute of Technology fur his thoughtful com-
ments on this topic. See his unpublished essay, "Wor!<ers and World Views."
In this chapter I elaborate on the postulate that there is reciprocity between the larger soci-
ety and bureaucratic institutions. It follows that in different national (or even subnational) cul-
tural settings, there will be manifest differences in bureaucratic organization.
3 These findings are drawn from Gabriel Almond and Sidney Verba, The Civic Culture
Notes
(Boston: Little, Brown, 1g65); and Michael Banton, The Policeman in the Community (London:
Tavistock, 1g64), cited in Elihu Katz and Brenda Danet, eds., Bureaucracy and the Public (New
York: Basic Books, 1973), p. 33 See their summary discussion of bureaucracy and culture, pp.
31-42
4 Katz and Danet, Bureaucracy and the Public, p. 34
5 The dynamics of the dialectic of expansion and contraction in public service benefits are
treated in Frances F. Piven and Richard Cloward, Regulating the Poor (New York: Pantheon,
1971). See also Michael Lipsky, Protest in City Politics (Chicago: Rand McNally, 1970), chap. 2;
Murray Edelman, Political Language (New York: Academic Press, 1977), chap. 3
6. To my knowledge few studies have inquired into public agency behavior under condi-
tions varying with agency need to attract clients. For one essay that addresses this consider-
ation, see Michael Lipsky and Morris I..ounds, "Citizen Participation and Health Care: Prob-
lems of Government Induced Participation," journal of Health Politics, Policy and Law, vol. 1,
no. 1 (Spring, 1976), pp. 85-111.
7 For illuminating discussions of the role of social welfare programs, broadly conceived,
in contemporary American society, see James O'Connor, The Fiscal Crisis of the State (New York:
St. Martin's, 1973); Piven and Cloward, Regulating the Poor; Ira Katznelson, "The Crisis of the
Capitalist City: Urban Politics and Social Control," in Willis Hawley and Michael Lipsky, eds.,
Theoretical Perspectives on Urban Politics (Englewood Cliffs, N.J.: Prentice-Hall, 1976), pp.
214-229.
8. See Jeffry Galper, The Politics of Social Services (Englewood Cliffs, N.J.: Prentice-
Hall, 1975).
9 Consider also Anthony Downs' discussion in "Why the Government Budget is Too Small
in a Democracy," World Politics, vol. 12, no. 4 (July, 1g6o), pp. 541-56:3.
10. On national variations in welfare benefit levels and administrative organization see
Harold Wilensky, The Welfare State and Equality: Structural and Ideological Roots of Public
Expenditures (Berkeley, Calif.: University of California Press, 1975).
11. On the symbolic significance of public policies see Edelman, Political Language.
12. I have stressed throughout the buffer role of street-level bureaucrats, especially in
chaps. 5, 6, and 9 For further discussion see Katznelson, "The Crisis of the Capitalist City," and
James D. Thompson, "Organizations and Output Transactions," in Katz and Dane!, eds.,
Bureaucracy and the Public, pp. 191-211.
13. Morris Janowitz directs attention to ways in which bureaucracies help shape the clients
with whom they later interact in Social Control and the Welfare State (New York: Elsevier,
1976).
14. Ibid., p. 105. Janowitz asserts that this is generally characteristic of service bureau-
cracies in welfare states: "Whether one is dealing with the format of public housing or with wel-
fare services associated with family assistance programs and community development, the
overall effect on the process of socialization is to separate and in fact isolate the clients from the
larger social structure and to seek to treat their needs in a very fragmented fashion. While these
programs have eliminated the stark misery of oppressive poverty and the fear of starvation, they
contain strong built-in limitations that thwart self-esteem and competence among recipients."
15. For approaches to politics that assume the ubiquity of conflict see Ralf Dahrendorf,
Class and Class Conflict in Industrial Society (Stanford, Calif.: Stanford University Press. 196g);
William Gamson, "Stable Unrepresentation in American Politics," American Behavioral Scien-
tist (November-Decem ber, 1g68), pp. 15-21. See also Lewis Coser, The Functions of Social
Conflict (New York: Free Press, 1g64).
16. Margaret Levi, "Poor People against the State," Review of Radical Political Economics,
vol. 6 (Spring, 1974), pp. 7&-79.
17. For one example of public employees seeking improved services fur citizens see the ef-
forts of the Service Employees International Union to obtain better patient care and treatment
facilities at Boston City Hospital. Boston Herald-American, May 25, 1978, p. 7
Notes
Chapter 13
1. See Murray Edelman, Political Language: Words That Succeed and Policies thot Fail
(New York: Academic Press, 1977), Chap. 4
2. For a natural experiment in which, because of a strike in 1g67, a city discovered whether
it needed the services of a type of street-level bureaucrat, see Arnold Weber, "Paradise Lost:
Or Whatever Happened to the Chicago Social Workers?" in Joseph Loewenberg and Michael
Moskow, eds., Collective Bargaining in Government (Englewood Cliffs, N.J.: Prentice-Hall,
1973).
3 Fred Hechinger, "Smaller Classes Found to Produce Subtle Changes," New York Times,
April 10, 1979. p. Cs. The relationship of class size to achievement may depend on the subject
taught. A rece!lt South Carolina experiment found signifieantly higher reading scores in classes
averaging 19.9 students compared to classes averaging 26.7 students. But no appreciable dif-
ferences were discovered in math scores. See New York Times, June 22, 1977, p. 35
4 Boston Globe, September 11, 1977. p. 10.
5 For the argument that work structure is more important than training in detennining
physicians' attitudes and the character of their practice see Eliot Freidson, Profession of Medi-
cine (New York: Dodd, Mead, 1974), chap. 5
Richard Weatherley reminds me in a letter that for current workers, training often serves the
latent function of providing a break from the job and is a fringe benefit contributing to worker
morale, whatever the effect on practice.
6. See John H. McNamara, "Uncertainties in Police Work: The Relevance of Police Re-
cruits' Backgrounds and Training," in David Bordua, ed., The Police (New York: John Wiley,
1967). pp. 16:3-252.
7 See Harold Wilensky, "The Professionalization of Everyone," American journal of Soci-
ology, vol. 70, no. 2 (September, 1g64); Amitai Etzioni, The Semi-Professions and their Organi-
zation (New York: The Free Press, 1g6g).
8. Usually, increased professionalization is associated with deference of the society toward
occupations, discretionary judgment, citizen trust, and reciprocal altruism, as well as the char-
acteristics mentioned here. However, in some cases the tenn simply denotes an occupational
group's increased adherence to accepted occupational nonns. Thus increased professionalism
among police may mean less discretion and greater confonnity to legal standards and nonns of
police conduct.
9 Deboralt Stone, "Professionals and Social Services," (unpublished paper, March, 1976);
Gideon Sjoberg et al., "Bureaucracy and the Lower Class," in Sociology and Social Re-
search, vol. so (1g66), pp. 325-337
10. On peer review in medicine see Freidson, Profession of Medicine, pp. 13711'.
11. Ronald Gross and Paul Ostennan, ed., The New Professionals (New York: Simon and
Schuster, 1972).
12. For a discussion of increased client participation in service bureaucracies (one generally
consistent with the arguments of this chapter), see Alan Gartner and Frank Reissman, The Ser~
vice Society and the Consumer Vanguard (New York: Harper and Row, 1974).
13. On the limitations of group practice in medicine relative to dominant professional
nonns, see Eliot Freidson, Doctoring Together: A Study of Professional Social Control (New
York: Elsevier, 1975).
Notes
Chapter 14
1. James W. Davis Jr. and Kenneth Dolbeare, Little Groups of Neighbors: The Selective
Service System (Chicago: Markham, 1968).
z. Martha Derthick, New Towns in-Town (Washington, D.C.: Urban Institute Press,
197Z).
3 Jeffrey Pressman and Aaron Wildavsky, Implementation (Berkeley: University of Cali-
fornia Press, 1973). For the first stirrings of the field of implementation studies, see Austin
Ranney, "The Study of Policy Content: A Framework for Choice," in Ranney, ed., Political
Science and Public Policy (Chicago: Markham, 1968), pp. 3-Zl.
4 C. Eugene Steurle, "Financing the American State at the Tum of the Century," in
W.E. Brownlee, ed., Funding the Modem American State, 1941-1gg6: The Rise and Fall of
the Era of Easy Finance (Cambridge: Cambridge University Press, 1996), pp. 409-444.
S Iris J. Lav, Elizabeth McNichol, and Robert Zahradnik, "Faulty Foundations: State
Structural Budget Problems and How to Fix Them" (Washington, D.C.: Center on Budget
and Policy Priorities, May 17, zoos). available at: http://www.cbpp.orglfiles/s-17-0Ssfp.pdf
(accessed December 10, zoog).
6. Contracting for services with for-profit agencies may be an entirely different matter.
The logic of profit-driven provision of services would predict that management give priority
to increasing caseloads and demanding that workers streamline and routinize interactions
with clients. See, for example, Janice Johnson Dias and Steven Maynard-Moody, "For Profit
Welfare: Contracts, Conflicts, and the Performance Paradox," Journal of Public Administra-
tion Research and Theory, vo!. 17, no. z (zoo7), pp. 18g--zu. Such pressures exist in the
public sector, to be sure, but perhaps not so overtly or legitimized by the rationale of the
market. The discussion of contracting for services draws on Steven Rathgeb Smith and Mi-
chael Lipsky, Nonprofits for Hire: The Welfare State in the Age of Contracting (Cambridge,
Mass.: Harvard University Press, 1993).
7 For a discussion on professionalism as an autonomous source of accountability in or-
ganizations, see Carolyn J. Hill and Laurence E. Lynn Jr., Public Management: A Three-Di-
mensional Approach (Washington, D.C.: CQ Press, zoog), pp. 3zz-3zs.
8. This section draws on Michael Lipsky, "Revenues and Access to Public Benefit," in
Jorrit de Jong and Cowher Rizvi, eds., The State of Access: Success and Failure of Democra-
cies to Create Equal Opportunities (Washington, D.C.: Brookings Institution Press, zooS),
PP 137-147.
g. Elizabeth McNichol and Nicholas Johnson, "Recession Continues to Batter State
Budgets; State Responses Could Slow Recovery" (Washington, D.C.: Center on Budget and
Policy Priorities, zoog), available at: http://www.cbpp.orglfiles/g-8-o8sfp.pdf (accessed De-
cember 10, zoog).
10. Jon Honeck, "The Governor's Announcement on Taxes" (Cleveland, Ohio: The Center
for Community Solutions, October 1, zoog), available at: http://www.communitysolutions
.corn/images/upload!resources/IncomeTaxStatementlooiOg.pdf (accessed December 19, zoog).
11. New York Times, October g, zoog, p. Azs.
1z. A staple of public opinion research is that people generally like what government
does, but tend to have a poor opinion of government as such. See, for example, Meg Bostrom,
"By, or for, the People? A Meta-analysis of Public Opinion of Government," (New York:
Demos, zoos). available at: http://www.demos.org!pubs/ByOrForthePeoplezooso4z6.pdf (ac-
cessed December 10, zoog).
13. Paul Pierson, "From Expansion to Austerity: The New Politics of Taxing and Spend-
ing," in Martin Levin et a!., eds., Seeking the Center: Politics and Policymaking at the New
Century (Washington, D.C.: Georgetown University Press, zoo1), p. 66.
14. The primary exception in the United States has been the transformation of the basic
welfare program for dependent children from an entitlement program to a time-limited as-
sistance program emphasizing work effort.
Notes
15. "Kaiser/Harvard Consumer Protection in Managed Care Smvey, December, 1997,"
Caring, vol. 17, no. 3 (March, 1998), pp. 1z-zo, available at: http://www.ncbi.nlm.nih.gov/
pubmed!101790ZO (accessed December 10, zoog). Generally, support for government initia-
tives to solve social and economic problems has been on the rise since zoo1, as has skepti-
cism toward elected officials. For a review before developments associated with the election
of President Obama and the economic collapse, see Trends in Political Values and Core At-
titudes: 1987-Z007 (Washington, D.C.: Pew Research Center for the Public & the Press,
March zz, zoo7), available at: http://people-press.orwreportl31z/trends-in-political-values
-and-core-attitudes-1987-ZOO? (accessed November zz, zoog).
16. Among other works, see Murray Edelman, The Symbolic Uses of Politics (Urbana:
University of Illinois Press, 1964).
17. For example, one can test whether providing a low income family with a housing
voucher improves its members' life chances. See James Rosenbaum and Stephanie DeLuca,
"Is Housing Mobility the Key to Welfare Reform?" (Washington, D.C.: Center on Urban and
Metropolitan Policy, The Brookings Institution, September, zooo), available at: http://www
.brookings.edu/-/media!Files/rc/reports/zooo/ogmetropolitanpolicy_rosenbaum/rosenbaum.
pdf (accessed October z3, zoog).
18. See Lizbeth Schorr, "Charities' Work Demands Flexible Evaluation," Chronicle of
Philanthropy, August zo, zoog, pp. 33ff.; also Michael Edwards, Small Change: Why Busi-
ness Won't Save the World (San Francisco: Barrett-Koehler, zoog).
19. In his inaugural address, President Obama emphasized that the role of government
should be approached pragmatically, and that government accountability is of the highest
priority. "The question we ask today is not whether our government is too big or too small,
but whether it works-whether it helps families find jobs at a decent wage, care they can af-
ford, a retirement that is dignified .... [T]hose of us who manage the public's dollars will be
held to account-to spend wisely, reform bad habits, and do our business in the light of
day-because only then can we restore the vital trust between a people and their govern-
ment." Obama Inaugural Speeches, January zo, zoog, available at: http://obamaspeeches.com
(accessed December 10, zoog).
zo. For a discussion that begins to treat the complexity of generalizing about account-
ability over the range of occupational roles and work environments that are encompassed by
the term street-level bureaucracy, see Peter Hupe and Michael Hill, "Street-Level Bureau-
cracy and Public Accountability,'' Public Administration vol. 85, no. z (zoo7), pp. z7g--zgg.
Zl. See pp. 3. xii, this volume. Some have even suggested that the book may be read as
unequivocally endorsing the discretionary behavior street-level bureaucrats exhibit. See, for
example, Janet Coble Vinzant and Lane Crothers, Street-Level Leadership: Discretion and
Legitimacy in Front Line Public Service (Washington, D.C.: Georgetown University Press,
ZOO?). P g.
zz. Evelyn Brodkin, "Bureaucracy Redux: Management Reformism and the Welfare
State," Journal of Public Administration Research and Theory, vo!. 17, no. 1 (zoo7), pp. 1-17.
z3. New York Times, April zg, zoog, p. A1g.
z4. See pp. 196--1g8, this volume.
z5. But even clerks can exercise discretion. See Saul Weiner et a!., "Rationing Access to
Care to the Medically Uninsured: The Role of Bureaucratic Front-Line Discretion at Large
Healthcare Institutions," Medical Care, vol. 4z, no. 4 (April, zoo4), pp. 306--31z.
z6. See Ian Taylor, "Discretion and Control in Education: The Teacher as Street-level
Bureaucrat," Educational Management Administration & Leadership, vo!. 35, no. 4 (zoo7),
pp. 555-57Z; Tony Evans and John Harris, "Street-Level Bureaucracy, Social Work and the
(Exaggerated) Death of Discretion," British Journal of Social Work, vol. 34 (zoo4), pp. 871-
895.
z7. Virginia Department of Motor Vehicles, "Virginia Performs," available at: http://
vaperforms.virginia.gov/agencyleveVsrc/ViewAgency.cfm ?agencycode= 154 (accessed October
z4, zoog).
zB. Celia Hagert, "Updating and Outsourcing Enrollment in Public Benefits: The Texas
Experience" (Austin, Tex.: Center for Public Policy Priorities, November, zoo6), available at:
http://www.cppp.orwfiles/3fCPPP_PrivReport_ %z8FS%zg.pdf (accessed October 11, zoog).
Notes
zg. On the hidden costs to citizens of efforts to save money through what are advertized
as essentially administrative reform, see Michael Lipsky, "Bureaucratic Disentitlement in So-
cial Welfare Programs," Social Service Review, vo!. 58 (1984), pp. 1-17.
30. Greg Marston, "Employment Services in an Age of E-Govemment," Information,
Communication & Society, vol. g, no. 1 (February zoo6), pp. 83-103.
31. David Landsbergen, "Screen Level Bureaucracy: Databases as Public Records,"
Government Information Quarterly, vo!. 21 (2004), p. 25.
32. In this section I draw on Michael Lipsky, "The Paradox of Managing Discretionary
Workers in Social Welfare Policy," in Michael Adler eta!., eds., The Sociology of Social Secu-
rity (Edinburgh: Edinburgh University Press, 1991), pp. 212-228.
33 Evelyn Brodkin, The False Promise of Administrative Reform (Philadelphia: Temple
University Press, 1986).
34 Jerry Mashaw, Bureaucratic Justice: Managing Social Security Disability Claims
(New Haven, Conn.: Yale University Press, 1983).
35 Robert Behn, Leadership Counts: Lessons for Public Managers from the Massachu-
setts Welfare, Training, and Employment Program (Cambridge, Mass.: Harvard University
Press, 1991). This account also draws on my experiences as a consultant with the Massachu-
setts Department of Public Welfare from 1983 to 1988.
The department leaders resisted evaluation through random assignment, despite consid-
erable pressure to do so, primarily for two reasons. First, they believed in placing people in
jobs, and thus considered it wrong to deny some citizens access to important services. Sec-
ond, they believed that it would not be possible to isolate a true control group, because so
many ET activities consisted of statewide publicity and collective efforts to assist welfare re-
cipients. See Behn's extensive discussion of ET's fit with the standards of random assignment
experimentation, chap. g.
36. Ibid., chap. 1.
37 For an account of a welfare agency that did not put comparable management tools
in place and failed in transforming workers' orientations, see Marcia K. Meyers et a!., "On
the Front Lines of Welfare Delivery: Are Workers Implementing Policy Reforms?" Journal
of Policy Analysis and Management, vo!. 17, no. 1 (1gg8), pp. 1-22. Soeren Winter and col-
leagues in many papers have explored on an empirical basis the contributions of manage-
ment initiatives, public policy dictates, and other interventions to bringing street-level bu-
reaucrats' performance in line with policy directives in Denmark See, for example, Peter J.
May and Soeren Winter, "Politicians, Managers, and Street-Level Bureaucrats: Influences on
Policy Implementation," Journal of Public Administration Research and Theory, vo!. 19, no.
3 (zoog), pp. 453-476.
38. For the origins of CompSTAT, see Jack Maple, The Crime Fighter: Putting the Bad
Guys Out of Business (New York: Random House, zooo).
39 In many articles Robert Behn seeks to identify the core elements of the CompSTAT
approach for a range of public management reforms that purport to build on the model. See,
e.g., "Designing PerformanceStat," Public Performance and Management Review, vol. 32, no.
z (December, zoo8), pp. 206-235
40. For another instance of shaping workers' behavior in the direction of generosity, see
Robert Garot, "Bias Forged through Suspicion: The Housing Gatekeeper Reconsidered," in
Stacy Bums, ed., Sociology of Crime, Law, and Deviance, vo!. 6 (Greenwich, Conn.: JAI
Press, zoos), pp. 77-104.
41. See p. 15, this volume.
42. Lisa Dodson, The Moral Underground: How Ordinary Americans Subvert the Un-
fair Economy (New York: New Press, zoog). This behavior has been described as "sabotage"
of agency policy in John Brehm and Scott Gates, Working, Shirking, and Sabotage: Bureau-
cratic Response to a Democratic Public (Ann Arbor: University of Michigan Press, 1997).
43 Steven Maynard-Moody and Michael Musheno, Cops, Teachers, Counselors: Stories
from the Front Lines of Public Service (Ann Arbor: University of Michigan Press, 2003), p.
24.
44 Michael Hill and Peter Hupe, Implementing Public Policy (London: Sage Publica-
tions, zooz), p. 27.
Notes
45 "The law's emphasis needs to shift from applying sanctions for failing to raise test
scores to holding states and localities accountable for making the systemic changes that im-
prove student achievement." "Joint Organizational Statement on No Child Left Behind
(NCLB) Act" (Boston, Mass.: FairTest, zoo4), available at: http://www.fairtest.or!joint%zo
statement%zocivil%zorights%zogrps%zo10-z1-04.html (accessed December 10, zoog).
46. To take an example with significant implications, in some schools serving Native
American students study units focusing on the history and culture of Native Americans-
subjects critical to helping native youth understand and appreciate their heritage-have been
dropped by schools giving priority to test results. Presentation by Hon. Ernie St. Germaine,
Honoring Nations Symposium, Cambridge, Mass., September 18, zoog.
47 Gerald Grant and Christine E. Murray, Teaching in America: The Slow Revolution
(Cambridge, Mass.: Harvard University Press, 1999), p. z15.
48. Dorothy Anagnostopoulos uses a street-level bureaucracy perspective to analyze
Chicago teachers' responses to a new accountability agenda in "The New Accountability, Stu-
dent Failure, and Teachers' Work in Urban High Schools," Educational Policy vol. 17, no. 3
(July zoo3), pp. zg1-316.
49 Linda Darling-Hammond and Milbrey Wallin McLaughlin, "Investing in Teaching
as a Learning Profession," in Darling-Hammond and Gary Sykes, eds., Teaching as the
Learning Profession (San Francisco: Jossey-Bass, 1999), pp. 376-411; see also Ron Haskins
and Susanna Loeb, "A Plan to Improve the Quality of Teaching in American Schools" (policy
brief, The Brookings Institution, Spring, zoo7), available at: http://www.brookings.edu/-/
media!Files/rc/papers/zoo7/spring_childrenfamilies_haskins/spring_childrenfamilies_
haskins.pdf (accessed November zg, zoog).
so. Mary E. Dilworth and Joseph A. Aguerrebere, "NCLB's Highly Qualified Teacher: A
Placeholder Definition," National Journal of Urban Education and Practice, vol. 1, no. z
(Fall, zoo7), pp. 111-135.
51. See Dilworth and Aguerrebere, p. 1Z7.
sz. Dilworth and Aguerrebere, p. 119.
53 See chap. 4
54 Letter to the legislature, Lewis H. Spence, Commissioner of the Department of So-
cial Services, April z4, zooz, p. 3
55 Lewis H. Spence, Letter to Fellow Employees, January z1, zoos, p. z.
56. Linda T. Kohn eta!., eds., To Err Is Human: Building a Safer Health System, Com-
mittee on Quality of Health Care in America (Washington, D.C.: National Academy Press,
zooo), available at: http://www.nap.edu/openbook.php?isbn=o3ogo68371 (accessed October
19, zoog). See pp. zog-z10, this volume.
57 On the relationship of citizen trust and experiences with street-level policy, see Bo
Rothstein, "The State and Social Capital: An Institutional Theory of Generalized Trust,"
Comparative Politics (July zooS), pp. 441-459.
58. See Margaret Levi, "A State of Trust," in Valerie Braithwaite and Levi, eds., Trust
and Governance (New York: Russell Sage Foundation, 1998), pp. 77-101.
59 Murray Edelman, Constructing the Political Spectacle (Chicago: University of Chi-
cago Press, 1988).
6o. See p. 5, this volume, and "U.S. Census of Government Employment by Geography
and by Government Function," March, zoo7.
266
INDEX
268
Index
Eisenstadt, S. N., 248n-250n Goodrich, C. H., 245n
Eligibility, 60--61 Goodwin, David, 260n
Ellis Island, 62 Government's role in society, increased
Emergencies, 136--39 criticism of, 214-215, 217-221
Emergency rooms, 33, 37, 43, 66; bias in, Great Britain, 180
109; and client education, 139; difficulties Great Recession (2008), 217
of performance measurement in, 168; Greenberg, Stanley, 250n
resource husbanding in, 125-126; role of Greenlick,~erwyn,245n
clerks in, 129--130; treatment of clients in, Gross, Ronald, 262n
85; triaging in, 106; waiting in, 98--99; see Groub~Barbara,239n-240n
also Hospitals Gypsies, 124-125
Emerson, Robert, 248n-249n, 253n
Employment agency, 51 Hahn, Harlan, 240n, 249n
Employment and Training Choices (ET), Halfway houses, 67
226--228,265n Hamilton, Edward K., 259n
Employment counselors, 74, 107--8, 166 Handler, Joel, 249n-250n, 255n
Employment programs, 11, 226--228 Hansen, Donald, 243n
English, Richard, 246n, 248n, 25ln, Hasenfeld, Yeheskel, 246n, 248n, 25ln, 253--
253n-255n 255n
Epstein, Irwin, 256n Hatry, Harry, 259n
Equal opportunity officers, 136 Hawley, Willis, 40, 239n, 246n-248n,
Etzioni, Amitai, 76, 243n, 246n, 248n, 250n, 250n-25ln,255n,26ln
257n, 262n Hayward, Nancy S., 259n
Exceptional children, 70 Head Start, 7
Health care, 32, 72, 207; home, 194-195
Farrar, Eleanor, 248n Health centers, 10; neighborhood control of,
Favoritism, 151-152 196; queue systems in, 96
Fire departments, 136 Health service careers, 3, 5, 206
Fiscal crisis, 172--179, 186, 240n; and Hechinger, Fred, 24ln, 262n
bureaucratic accountability, 159; and the Henry, Jules, 254n
firing of public employees, 7--8, 260n Heumann, ~ilton, 245n
Fiscal stringency, 214, 216--217 Hill, ~ichael, 231
Fogelson, Robert, 245n Hirschman, Albert 0., 52, 247n
Food stampJ.rogram, 45, 89, 246n Hollingsworth, ~ary Jane, 249n-250n, 255n
Ford, Cera! , 246n Holzer, ~ark, 259n
Forer, Lois, 239n, 254n Home health care, 194-195
Forms, 120-21, 150 Home visits, 120, 150
For-profit agencies, public service contracts Hospitals, 114, 203; emergency procedures
with, 263n in, 136; and the ideology of benign
Freidson, Eliot, 112, 141, 248n-249n, 254n, intervention, 119; inquiry policies in, 209--
256n-258n, 262n 210; protocols for medical priorities in,
Fuchs, Victor, 240n l98;VA,20-21,23, l29,24ln;voluntary,
Funding, 92--93 109; see also Emergency rooms
Hosticka, Carl, 120, 240n, 243n, 245n,
Galper, Jeffry, 132, 239n-240n, 253n, 256n, 253n-255n, 257n
26ln Housing, 14,21-22,88, 135, 254n;
Gamson, William, 26ln discriminatory decision making in public,
Gartner, Alan, 240n, 262n 108; and emergency status, 137; inspection
Gaylin, Willard, 254n of, 52 (see also Housing inspectors);
Ginsburg, Sigmund, 242n integration of public, 46; and rent strikes,
Gintis, Herbert, 240n 63--64; role of applications clerk in public,
Glaser, Barney, 253n 129; selective aid to applicants for public,
Goal clarification, 164 65,97,137, l50-l5l;tenantmanagement
Coffman, Erving, 66, 154, 248n-249n, 254n, of public, 193
258n Housing inspectors, 52, 101, 127-128; see
Goldbricking, 143 also Building inspectors; Housing
Goldstein, Herman, 247n Human relations staffs, 135--136
Index
Human services, 27, 192-211; see also Public Landau, Martin, 40, 246n
service workers Landlords, 135
Hupe, Peter, 231 Lasswell, Harold, 251n
Law enforcement personnel; see Police
Idealism; see Altruism officers
Ideology, 147-148 Law practice, 202, 206; see also Lawyers
Immigration regulations, 113 Lawyers, 19, 65, 98, 206; accountability of,
Implementation studies, 213--214 163; client control by, 120--122, 124;
Income inequalities, 7,54-55 insulation of, from client demands, 100;
Inflation,39, 170,176 perceived need for, 91; practice standards
Information, 90--94 of, 202-203; psychological sanctions
Information technology, 224-225 imposed by, 94; role conflicts of, 42, 162;
Integration,21-22,46,246n and specialization, 146; training of, 71;
Isolation, work in, 203, 206, 208 work conditions of, 30--31, 34, 36; see also
Israel, 62, 108 Legal services lawyers; Legal services
programs; Public service lawyers
Jackson, Kenneth T., 245n Lazarus,Richard,251n
Jacob, Herbert, 244n, 246n, 260n Lebeaux, Charles, 257n
Jacobs, Paul, 243n Legal services lawyers: client control by, 120--
Jacobson,Lenore,240n,249n 122, 124; routines of, 150; support for, 205;
Jails, 5; see also Correctional facilities; Prisons see also Lawyers; Legal services programs;
Janowitz, Morris, 261n Public service lawyers
Japha, Anthony, 257n Legal services programs, 96, 99;
Jargon, 90 neighborhood control of, 196; small group
Job satisfaction, 140 decision making for, 207; see also Lawyers;
Job stresses, 141-144; and defenses against Legal services lawyers; Public service
discretion, 149-150; and specialization, lawyers
146; see also Role conflicts Leibowitz, Hanna B., 260n
Job training counselors, 153 Levi, Margaret, 240n, 249n, 261n
Job training programs, 49, 227-228, 265n Levin, Martin, 44, 243n-244n, 246n-247n
Judges, 3, 7, 97, 145; and advocacy, 74; basis Lewin, David, 239n
of sentencing by, 109, 111-112, 124-125; Libraries,5,58, 102
controlling administrative law judges, 226; Liebow, Elliot, 252n
and the courtroom setting, 118; decision Life insurance industry, 84
making by, 128; insulation of, from client Lindblom, Charles, 251n
demands, 100; job reconceptualizations by, Lindzey, Gardner, 246n
150--151; as policy makers, 13, 19-20, 25, Lipsky, Michael, 239n-240n, 245n-246n,
85; professional identity of, 190; 249n-250n,252n,256n-257n,259n,2 61n
psychological sanctions from, 94; rubber Little City Halls programs, 35, 92
stamping by, 129-131; and their Loewenberg,J.Joseph,240n,262n
nonvoluntary clients, 58, 62, 64, 68; work Long Island Expressway, 33, 38, 199
conditions of, 30, 34, 44; see also Courts Los Angeles, 32
Juryimpaneling,98 Lounds, Morris,246n,252n,261n
Luckman, Thomas, 248n-249n, 251n
Kah, Marianne Stein, 239n Lyford, Joseph, 243n
Katz,Eiihu,248n-2150n,256n-258n, 261n
Katznelson, Ira, 240n, 248n, 261n Mainstreaming, 77
Keith-Lucas, Alan, 249n-250n, 252n Managers, 18--25, 242n
Kelner, Bernard G., 244n, 255n Mandatory sentencing legislation, 151
Kettleson, Jeanne, 246n Manuals, 162
Key, V 0., Jr., 180, 260n March, James, 242n, 251n, 257n, 260n
Kirkham, George, 32, 244n Margolis, B. L., 244n, 247n
Kirp, David, 245n, 250n, 253n, 255n Marriage counselors, 130
Kitsuse, John 1., 247n, 250n, 252n Marston, Greg, 224
Kozol, Jonathan, 258n Massachusetts, 177; alternatives to
Krugman, Paul, 217 incarceration in, 19-20; special education
Kuh, Richard, 22-23 issues in, 74, 127
Index
Mayhew, Leon, 252n Nonprofit organizations, public seiVices from,
Maynard-Moody, Steven, 230 215-216
McCorry, Jesse, 260n Nordlinger, Eric, 242n, 245n, 252n,
McNamara, John H., 244n, 256n, 262n 257n-258n
Mechanic, David, 24, 242n, 249n, 254n Nurses, 19, 110, 201-202; see also Medical
Media, emphasis on government failures over practice
successes, 218 Nutt, Thomas E., 255n
Medicaid, 88
Medical examiners, 130 Obama, Barack, 220, 264n
Medicalization of social problems, 12, 148 O'Connor, James, 240n, 26ln
Medical practice, 120, 160, 202, 258n; neglect Olson, David J., 245n
and abuse in, 56; peer review in, 262n; see Organizations: accountability in, 159--164;
also Doctors; Hospitals; Nurses goal clarification in, 164-165; performance
Medical review boards, 160 measures in, 165-169; and policy making,
Medicare, 45, 69, 89, 225 82-86; see also Bureaucracies
Mediocrity, 144, 169 Orren, Karen, 25ln
Mental health centers, 33-34, 92-93, 96; Osterman, Paul, 262n
client differentiation in, 112; client-staff Outsourcing of public seiVices, 215-216, 224
interaction in, 94
Mental health workers, 85 PapeiWork, 163, 244n, 258n; and use of
Mental hospitals, 66-67, 78, 85, 154; under forms, 120-121, 150
budget freeze, 177; commitment to, 130 Paraprofessionals, 116, 171, 194, 245n; legal,
Mental illness, 69 198
Mental retardation, 77 Parole personnel, 12, 110
Mercer, Blaine, 244n, 255n Paternalism, 79, 197
Merit increases, 206 Peer review, 203, 262n; and peer support,
Merton, Robert, 249n 206-207,209
Messinger, Sheldon, 243n Pension rights, 143
Methadone, 22 Performance measures, 48-53; and
Mileski, Maureen, 112, 244n, 254n-255n accountability, 162-163, 165-169, 199; and
Milloy, Ross, 259n merit increases, 206
Minorities, 10, 135, 204 Perlman, Robert, 100, 244n, 253n,
Morris, FrankL., Sr., 24ln 256n-257n
Moskow, Michael H., 240n, 262n Perry, David, 24ln, 256n
Moynihan, Daniel P., 246n Personality development, 66-70
Museums, 102 Peter Principle, 126
Musheno, Michael, 230 Physicians; see Doctors
Pierson, Paul, 217
National Association for the Advancement of Piliavin,IIVing,244n,247n,249n,254n
Colored People (NAACP), Ill Pilot programs, 187
National Board for Professional Teaching Piven, Frances Fox, 10, 240n, 252n, 256n,
Standards (NBPTS), 232 260n-26ln
New York City, 7, 134-135, 137, 167; Board Police officers, 239n-240n, 257n;
of Education, Ill; 911 system in, 35; accountability of, 159, 163-169; advocacy
police in, 32, 240n; rent-strike movement by, 74; as agents of social control, ll-12;
in, 63--64; and the Stavisky bill, 240n; alienation of, 78-79; allegations of
teacher staff reductions in, 176-177; brutality by, 135; altruism of, 185;
Transit Authority in, 171; welfare rolls in, American, compared to British, 180; and
22,91,103-104 bias, 108-111, 114; and citizen patrols,
Niederhoffer, Arthur, 247n, 255n 194; and citizen renew boards, 124; client
911 system, 35, 92, 129 control by, 122-125, 136; CompSTAT
Nivola, Pietro, 242n, 247n, 249n, 253n, 255n initiative, 228-229; as deliverers of
Nixon Administration, 164 government policy, 3-6; and department
No Child Left Behind (NCLB), 231-233, reform, 228-229, 247n; emergency
266n procedures of, 136-137; and the fiscal
No-fault case review for CPS, 235 crisis, 175; incompetent, 187; insulation of,
Noncooperation, 17 from client demands, 100; manipulation of
Index
Police officers (cont.) alienation of, 75-80; and bureaucratic
statistics by, 259n; performance measures reform, 192-211; client control by, 128--
for, 50-52; as policy makers, 13-15, 18--19, 139; client differentiation by, 105-116; and
25, 84-86; private goal definition by, 145- the client-processing mentality, 140-156;
146; professional norms of, 190; reality- collective bargaining by, for client needs,
based training for, 209; resource 190-191; deliverers of government policy,
husbanding by, 125-126; role conflicts of, 3-4, 8--12; evolving role of, 212-214,236-
40-48,73-75,105, 115,246n-247n; 237; and fiscal crisis, 172-179; husbanding
rubber-stamping of recommendations by, of resources by, 125-128; and the myth of
129-131; self-image of, 82, 114; service service altruism, 71-80; noncooperation
rationing by, 87-89, 92, 95, 102-103, 106; of, 17, 23-25; performance measures for,
small group decision-making by, 207; 165-169; as policy makers, 13-25, 144;
support for, 205-206, 210; and their reductions in during recession, 217; role
nonvoluntary clients, 54, 62-63. 66; and conflicts of, 188--191; routines and
their partners, 203-204; work conditions simplifications of, 83-86; service rationing
of, 30--32, 34, 39; work rewards of, 75 by, 87-104; and societal values, 180-188;
Police review boards, 124, 135, 160 supporting idealism in, 204-211; work
Policy making, by street-level bureaucrats, conditions of, 27--39, 81-83; see also
13-25,84-85,207,221-2 23 Street-level bureaucrats
Post office, 88 Pynoos,Jon,242n, 249n
Preschool programs, 7
Pressure specialists, 133-135 Quality control in welfare administration,
Pretrial diversion programs, 20 225-226
Principals, 242n Queuing, 95-99
Prisons, 5-6,57, 114, 154; alternatives to, Quinney, Richard, 240n, 254n
19-20; client control in, 117, 119, 256n; Quitting, 17
and corrections policy, 12; policy making
in, 13-14; and stigma, 69; work conditions Racism, 111, 115, 181-182,213
in, 78; see also Correctional facilities Raffel, Jeffrey, 246n
Privacy, 64, 118 Rainwater, Lee, 251n
Private sector, provision of public services by, Rape, 110-111
215-216,224 Rationingservices,87-10 4, 173;through
Probation officers, 20, 125; weight of client differentiation, 105-116
recommendations by, 25, 130-131 Receptionists, 128--129, 198, 254n
Productivity, 170-172 Record keeping, 163, 258n; see also
Professionalism, 189-190, 262n; ideology of, Paperwork
147-148; and the myth of service altruism, Red tape, 63
71-80; prospects and problems of, 201-211 Reference groups, 45-48, 57, 81; promoting
Prosecutors, 150-151 client impact on, 208
Prottas, Jeffrey, 242n-243n, 248n, Referrals, 132-133
252n-253n,255n Reich, Charles, 240n
Psychiatrists, 20, 31, 44; client differentiation Rein, Martin, 240n, 246n
by, 111 Reiss, Albert, Jr., 136, 241n, 247n, 252n, 256n
Psychological sanctions, 93-94 Reissman, Catherine Kohler, 96, 244n-245n,
Psychologists, 5, 77, 111, 113 253n
Public defenders, 74, 101, 112, 145; Reissman. Frank, 240n, 262n
rubberstamping by, 130; working Rent-strike movement, 63-64
conditions of, 29, 34 Research standards for public services
Public housing; see Housing evaluation,219-220
Public service lawyers, 3-7; altruism of, 185; Resources, deployment of: and emergency
bias among, 113; as teachers of the client procedures, 138; husbanding of, 125-128;
role, 60-62, 69; who quit, 143; working by public-service workers, 167; results of
conditions of, 29-30, 34, 36; see also added, 199-200
Lawyers; Legal services lawyers; Legal Responsiveness of public services, 229
services programs Retirement, 143
Public service workers: accountability of, Review boards, 124, 135, 160
159-165, 170-172; advocacy by, 72-75; Richardson, James, 245n
272
Index
Riley, Judy, 250n, 255n, 258n Semi-professionals, 194; see also
Rist, Ray C., 67-68, 249n, 255n Paraprofessionals
Rockefeller, Nelson, 22 Service Employee International Union, 261n
Rogers, David, 251n Service vouchers, 193-194
Role conflicts, 40-48, 105, ll5; and advocacy, Settings, ll7-120
73-75;forpollce,246n-247n Sharkansky, Ira, 251n
Rosenthal, Robert, 240n, 249n Silver, Allan, 240n, 245n
Rosett, Arthur, 257n Silver, Carol Ruth, 245n
Rossi, Peter, 254n, 258n Simon, Herbert, 242n, 251n, 257n
Roth,Jullus,85,248n, 251n,253n,255n Simplifications, 83-86, 251n; and
Routines, 83-86, 251n; and bureaucratic bureaucratic reform, 199
reform, 199; and client control, 121-125; Sjoberg, Gideon, 262n
consequences of, 133-136, 139; pollee, Skolnick, Jerome, 249n-250n
120, 122-125;andservicerationing,99- Slowdowns, 143, 175
104; worker attachment to, 149 Snyder, John A., 255n
Roxbury Multi-Service Center, 100, 132 Social security benefits, 91-92
Rubber-stamping, 128-131 Social Security Disability Insurance (SSDI)
Rubinstein, Jonathan, 242n, 247n, 252n, 255n program, 226
Ryan, William, 257n Social workers, 3-7, 77, 105, 150; altruism of,
185; client conceptions of, 155;
Sabel, Charles, 260n discretionary roles of, 197,221-229, 233-
Salaries, 5-7 236; incompetent, 187; insulation of, from
Sanitation services, 167, 173 client demands, 100; paternalism of, 79-
Sarbin, Theodore, 246n-247n 80; performance measures for, 50; as
Savas, E. S., 242n, 247n policy makers, 19-20; professionalization
Scheff, Thomas, 243n of, 190, 201-202; reality-based training for,
Schelling, Thomas, 248n 209-210; reporting required of, 182; role
Schild, E. 0., 256n-257n conflicts of, 162, 183; rubber-stamping of
School counselors, 73 recommendations by, 13<f--131; site visits
School psychologists, 62 by, 120; and specialization, 146; and their
Schools, 98, 146-147; accountability in, 159, involuntary clients, 57, 63, 65; work
165; bias in, 108, 1ll-ll4; client control conditions of, 3a--31, 36
in, ll9, 123-124; decentralization of, 160; Sornoff, Paula, 241n, 256n
Denver Plan for, 200; expulsion from, 88; Special education, 74, 77-78, 109, 127;
function of settings in, ll8; goal conflicts function of, 135; and issues of
in, 40, 42-44, 46, 165; the medical model specialization, 147
in, 148; neighborhood control of, 196; Specialization, 77-78; agency function of,
nonvoluntary clients of, 54-56; pressure 146-147; and fiscal crisis, 175; and
specialists in, 133; small group decision inequitable distribution of practitioners,
making in, 207; socialization function of, 202
la--ll; specialization in, 77, 178; tracking Spence, Harry, 233-234
in, 63, 67-70, 106, ll2, 130; use of Status, 202, 206
community resources by, 194; see also Stavisky bill, 240n
Education; Special Education; Teachers Stealing, 17
Schultz, Stanley, 245n Steiner, Gilbert, 240n, 246n, 252n
Schwartz, Barry, 244n, 252n-253n Stereotypes, ll5, 140, 142, 155
Schwartz, Charlotte Greene, 256n Stevenson, Keith, 241n
Schwartz, Morris, 256n Stigma,68-70,91-92, 182,184
Schwartz, Richard, 249n-250n Stinchcombe,Artbu~260n
Scott, Robert, 254n Stone, Clarence, 258n
Screening, 128-129 Stone,Deborah,245n,250n, 262n
Sears, David, 256n Strauss, Anselm, 253n
Secretaries, 128-129 Street-level bureaucrats: accountability to
Seidman, David, 247n, 259n policy, 221-229; as agents of social control,
Selective Service System, 84, 214, 256n 8-12; conclusion, 236-237; contradictory
Self-actualization, 250n tendencies of, 188-189; defined, 3; as
Self-images, 66-67, 82, ll4, 183 deliverers of government policy, 3-4;
273
Index
Street-level bureaucrats (cont.) Trees, Dennis, 255n
evolving policy environment, 212--221; Triage, 106-107, 118, 138
goal conflicts of, 40-48; investing in, 229- Turnover, 143, 174
236; nonprofit organizations, 216; and
nonvoluntary clients, 54--59; performance Ubell, Earl, 244n
measures for, 48-53; policy making by, Uniforms, 118
13-25, 84--85,207, 221-223; scope of Unions, 7, 17, 24; and bureaucratic reform,
services provided by, 4-8; shaping 206, 210; and educational voucher
performance of, 221-229; and the social systems, 93; efforts by, for clients' needs,
construction of clients, 59-70; work 80, 190--191
conditions of, 27-39; see also Public Universities, 39, 92--93, 208
service workers; individual occupations Upward Bound program, 49-50, 107
Sudnow, David, 112, 253n-255n Urban renewal programs, 8
Supervisors, 242n U.S. Census Bureau, 5
Support services, 31
Sweet, Donald H., 260n Van Hom, Carl, 24ln-242n
Sykes, Gresham, 24ln, 244n, 248n VanMaanen, John, 244n, 255n, 260n
Van Meter, Donald, 24ln-242n
Teachers,93, 105,107-108, 198,247n; Verba, Sidney, 260n
accountability of, 163, 166-169; as agents Veterans Administration benefits, 92
of social control, 11; alienation of, 78-80; Veterans Administration hospitals, 20--21, 23,
altruism of, 185; client conceptions of, 129,235,24ln
155, 254n; client control by, 120, 122, Violence, 31-32, 110--111, 135; and police
124; client differentiation by, 110--111, public image, 123
113, 152--153; collective bargaining for, Voucherproposals, 93, 160,193-194
for better classroom conditions, 190--191;
as deliverers of government policy, 3-5, VVagefreezes, 175--176
7-8; under the Denver Plan, 200; VVaiting, 89, 93; in queues, 95--99; and triage,
discretionary roles of, 197; and the fiscal 106
crisis, 175--177; incompetent, 187; VVar on poverty, 11, 41
paternalism of, 79-80; performance VVaskow, Arthur, 245n, 250n
measures for, 50-53, 232--233; as policy VVaste, 17
makers, 13, 15--16, 18-19, 85; private goal VVeatherley, Richard, 244n, 249n-250n, 255n,
definition by, 145--146; professionalization 257n,259n,262n
of, 190, 201-202; reporting required of, VVeber, Arnold, 262n
182; role conflicts of, 162, 183, 186-187; VVeinberg, Martha VVagner, 242n-243n
rubber-stamping by, 130; self-image of, VVeiss, James, 245n
82, 114; service rationing by, 89-90, 94; VVelfare agencies, 10, 46, 88, 139;
socializing role of, 61--63, 67; and accountability in, 159, 164--165; and client
specialization, 77; support for, 205--206, organizations, 119; discretionary options
210; and their nonvoluntary clients, 57- in, 149, 197,222-223;emer gency
58, 61--63; training of, 71; who quit, 143; procedures in, 136; in New York City, 22,
work conditions of, 30, 32, 36-37, 39; see 91, 103-104; nonvoluntary clients of, 54-
also Education; Schools 55, 64, 66, 69; policy making in, 14; and
Teaching machines, 198 program goal conflicts, 41-43;
Team approach to CPS case work, 235 psychological burdens borne by clients of,
Tenure system, 93, 143 93-95, 104; quality control in, 225--226;
Thayer, Frederick, 250n rationing of services by, 90; resource
Thompson, Elizabeth J., 254n, 258n husbanding in, 126-127; specialization in,
Thompson, James D., 166-167, 248n, 255n, 146; statistics generated by, 51; see also
258n,26ln VVelfare workers
Thompson, Victor, 25ln, 257n VVelfare rights movement, 63, 149
Time, 89-90 VVelfare state, 8, 11-12, 183; contradictions
Tracking, 106, 112; effects of, 63, 67-70; and in, 189
rubber-stamping, 130 VVelfare workers, 4; case presentations by,
Trauma-team personnel, 109-110 134; client control routines of, 124-125,
274
Index
150; discretion controls for, 225--226, Wilson, James Q., 229, 241n, 244n-247n,
265n; failure of, to differentiate among 250n-251n,257n-258n
clients, 122; response to 1996 reforms, Withdrawal behaviors, 142-144, 257n
222-223; self-image of, 82, 114; Wolf, June Grant, 252n
transforming orientations of, 226-228; see Wolff, Robert P., 240n
also Welfare agencies Women's rights movement, 135
Werthman, Carl, 244n, 249n, 254n Wynne, Edward, 258n
Westley, William, 244n
Wheeler, Stanton, 247n, 250n, 254n Yates, Douglas, 243n
White, Leonard D., 260n
Whitney, Richard, 256n Zald, M., 256n
Wildavsky, Aaron, 257n Zimmerman, Don, 243n-244n
Wilensky, Harold, 257n, 261n-262n Zoos, 102
Willemain, Thomas, 241n, 244n Zubkoff, Michael, 256n
275
STREET-LEVEL BUREAUCRACY
AWARDS FOR THE PREVIOUS EDITION
Winner of the 1980 C. Wright Mills Award, Society for the Study of Social Problems
Winner of the 1981 Gladys M. Kammerer Award, American Political Science Association
Winner of the 1999 Aaron Wildavsky Enduring Contribution Award of the Policy Studies Organization
First published in 1980, Street-Level Bureaucracy received critical acclaim for its insightful study
of how public service workers, wielding considerable discretion in how to execute their jobs,
function as policy decisionmakers. Three decades later, public urgency to bolster the availability
and effectiveness of healthcare, social services, education, and law enforcement has intensified,
making Street-Level Bureaucracy more relevant now than ever. In this thirtieth anniversary
expanded edition, Michael Upsky revisits the territory he mapped out in the first edition to reflect
on significant policy developments and show that street-level bureaucracies can be improved and
work in public service fields can be rewarding.
Street-Level Bureaucracy examines how discretionary services, work conditions, and work
practices interact to influence client outcomes. Street-level bureaucrats-from teachers and police
officers to social workers and legal-aid lawyers-interact directly with the public and so represent
the frontlines of government policy. Upsky argues that these relatively low-level employees in
human service agencies labor under huge caseloads, ambiguous agency goals, and inadequate
resources. When combined with substantial discretionary authority and their ability to interpret
policy on a case-by-case basis, the difference between government policy in theory and policy in
practice can become vast.
This seminal study tells a cautionary tale of how decisions made by overburdened workers
in underfunded government agencies translate into ad-hoc policy changes impacting peoples'
lives and life opportunities. This expanded edition of Street-Level Bureaucracy underscores that,
despite its challenging nature, street-level work can be a conduit for, rather than a barrier to,
providing services to citizens.
ISBN-13: 976-0-67154-544-2
ISBN-10 : 0-67154-544-b
RUSSELL SAGE FOUNDATION
New York, New York I www.russellsage.org
10871111 f